Instructions to use barha/granite-switch-4.0-350m-demo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use barha/granite-switch-4.0-350m-demo with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="barha/granite-switch-4.0-350m-demo")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("barha/granite-switch-4.0-350m-demo", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use barha/granite-switch-4.0-350m-demo with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "barha/granite-switch-4.0-350m-demo"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "barha/granite-switch-4.0-350m-demo",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/barha/granite-switch-4.0-350m-demo

SGLang

How to use barha/granite-switch-4.0-350m-demo with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "barha/granite-switch-4.0-350m-demo" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "barha/granite-switch-4.0-350m-demo",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "barha/granite-switch-4.0-350m-demo" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "barha/granite-switch-4.0-350m-demo",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use barha/granite-switch-4.0-350m-demo with Docker Model Runner:
```
docker model run hf.co/barha/granite-switch-4.0-350m-demo
```

granite-switch-4.0-350m-demo / compose_report.json

barha

granite-switch-4.0-350m 3-adapter aLoRA demo: cti-technique-mapping + text-to-json + genai-attack-vector

a6517e6 verified 2 months ago

Raw

History Blame Contribute Delete

194 kB

	{
	"metadata": {
	"timestamp": "2026-06-16T07:26:03.523574"
	},
	"base_model_mapping": [
	{
	"source": [
	"model.embed_tokens.weight"
	],
	"target": "model.embed_tokens.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.0.input_layernorm.weight"
	],
	"target": "model.layers.0.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.0.post_attention_layernorm.weight"
	],
	"target": "model.layers.0.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.0.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.0.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.0.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.0.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.0.self_attn.o_proj.weight"
	],
	"target": "model.layers.0.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.1.input_layernorm.weight"
	],
	"target": "model.layers.1.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.1.post_attention_layernorm.weight"
	],
	"target": "model.layers.1.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.1.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.1.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.1.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.1.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.1.self_attn.o_proj.weight"
	],
	"target": "model.layers.1.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.2.input_layernorm.weight"
	],
	"target": "model.layers.2.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.2.post_attention_layernorm.weight"
	],
	"target": "model.layers.2.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.2.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.2.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.2.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.2.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.2.self_attn.o_proj.weight"
	],
	"target": "model.layers.2.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.3.input_layernorm.weight"
	],
	"target": "model.layers.3.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.3.post_attention_layernorm.weight"
	],
	"target": "model.layers.3.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.3.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.3.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.3.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.3.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.3.self_attn.o_proj.weight"
	],
	"target": "model.layers.3.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.4.input_layernorm.weight"
	],
	"target": "model.layers.4.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.4.post_attention_layernorm.weight"
	],
	"target": "model.layers.4.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.4.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.4.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.4.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.4.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.4.self_attn.o_proj.weight"
	],
	"target": "model.layers.4.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.5.input_layernorm.weight"
	],
	"target": "model.layers.5.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.5.post_attention_layernorm.weight"
	],
	"target": "model.layers.5.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.5.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.5.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.5.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.5.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.5.self_attn.o_proj.weight"
	],
	"target": "model.layers.5.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.6.input_layernorm.weight"
	],
	"target": "model.layers.6.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.6.post_attention_layernorm.weight"
	],
	"target": "model.layers.6.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.6.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.6.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.6.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.6.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.6.self_attn.o_proj.weight"
	],
	"target": "model.layers.6.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.7.input_layernorm.weight"
	],
	"target": "model.layers.7.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.7.post_attention_layernorm.weight"
	],
	"target": "model.layers.7.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.7.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.7.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.7.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.7.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.7.self_attn.o_proj.weight"
	],
	"target": "model.layers.7.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.8.input_layernorm.weight"
	],
	"target": "model.layers.8.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.8.post_attention_layernorm.weight"
	],
	"target": "model.layers.8.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.8.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.8.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.8.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.8.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.8.self_attn.o_proj.weight"
	],
	"target": "model.layers.8.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.9.input_layernorm.weight"
	],
	"target": "model.layers.9.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.9.post_attention_layernorm.weight"
	],
	"target": "model.layers.9.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.9.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.9.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.9.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.9.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.9.self_attn.o_proj.weight"
	],
	"target": "model.layers.9.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.10.input_layernorm.weight"
	],
	"target": "model.layers.10.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.10.post_attention_layernorm.weight"
	],
	"target": "model.layers.10.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.10.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.10.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.10.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.10.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.10.self_attn.o_proj.weight"
	],
	"target": "model.layers.10.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.11.input_layernorm.weight"
	],
	"target": "model.layers.11.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.11.post_attention_layernorm.weight"
	],
	"target": "model.layers.11.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.11.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.11.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.11.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.11.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.11.self_attn.o_proj.weight"
	],
	"target": "model.layers.11.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.12.input_layernorm.weight"
	],
	"target": "model.layers.12.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.12.post_attention_layernorm.weight"
	],
	"target": "model.layers.12.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.12.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.12.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.12.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.12.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.12.self_attn.o_proj.weight"
	],
	"target": "model.layers.12.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.13.input_layernorm.weight"
	],
	"target": "model.layers.13.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.13.post_attention_layernorm.weight"
	],
	"target": "model.layers.13.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.13.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.13.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.13.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.13.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.13.self_attn.o_proj.weight"
	],
	"target": "model.layers.13.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.14.input_layernorm.weight"
	],
	"target": "model.layers.14.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.14.post_attention_layernorm.weight"
	],
	"target": "model.layers.14.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.14.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.14.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.14.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.14.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.14.self_attn.o_proj.weight"
	],
	"target": "model.layers.14.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.15.input_layernorm.weight"
	],
	"target": "model.layers.15.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.15.post_attention_layernorm.weight"
	],
	"target": "model.layers.15.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.15.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.15.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.15.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.15.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.15.self_attn.o_proj.weight"
	],
	"target": "model.layers.15.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.16.input_layernorm.weight"
	],
	"target": "model.layers.16.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.16.post_attention_layernorm.weight"
	],
	"target": "model.layers.16.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.16.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.16.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.16.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.16.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.16.self_attn.o_proj.weight"
	],
	"target": "model.layers.16.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.17.input_layernorm.weight"
	],
	"target": "model.layers.17.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.17.post_attention_layernorm.weight"
	],
	"target": "model.layers.17.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.17.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.17.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.17.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.17.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.17.self_attn.o_proj.weight"
	],
	"target": "model.layers.17.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.18.input_layernorm.weight"
	],
	"target": "model.layers.18.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.18.post_attention_layernorm.weight"
	],
	"target": "model.layers.18.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.18.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.18.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.18.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.18.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.18.self_attn.o_proj.weight"
	],
	"target": "model.layers.18.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.19.input_layernorm.weight"
	],
	"target": "model.layers.19.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.19.post_attention_layernorm.weight"
	],
	"target": "model.layers.19.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.19.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.19.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.19.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.19.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.19.self_attn.o_proj.weight"
	],
	"target": "model.layers.19.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.20.input_layernorm.weight"
	],
	"target": "model.layers.20.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.20.post_attention_layernorm.weight"
	],
	"target": "model.layers.20.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.20.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.20.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.20.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.20.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.20.self_attn.o_proj.weight"
	],
	"target": "model.layers.20.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.21.input_layernorm.weight"
	],
	"target": "model.layers.21.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.21.post_attention_layernorm.weight"
	],
	"target": "model.layers.21.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.21.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.21.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.21.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.21.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.21.self_attn.o_proj.weight"
	],
	"target": "model.layers.21.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.22.input_layernorm.weight"
	],
	"target": "model.layers.22.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.22.post_attention_layernorm.weight"
	],
	"target": "model.layers.22.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.22.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.22.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.22.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.22.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.22.self_attn.o_proj.weight"
	],
	"target": "model.layers.22.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.23.input_layernorm.weight"
	],
	"target": "model.layers.23.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.23.post_attention_layernorm.weight"
	],
	"target": "model.layers.23.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.23.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.23.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.23.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.23.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.23.self_attn.o_proj.weight"
	],
	"target": "model.layers.23.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.24.input_layernorm.weight"
	],
	"target": "model.layers.24.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.24.post_attention_layernorm.weight"
	],
	"target": "model.layers.24.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.24.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.24.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.24.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.24.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.24.self_attn.o_proj.weight"
	],
	"target": "model.layers.24.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.25.input_layernorm.weight"
	],
	"target": "model.layers.25.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.25.post_attention_layernorm.weight"
	],
	"target": "model.layers.25.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.25.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.25.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.25.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.25.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.25.self_attn.o_proj.weight"
	],
	"target": "model.layers.25.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.26.input_layernorm.weight"
	],
	"target": "model.layers.26.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.26.post_attention_layernorm.weight"
	],
	"target": "model.layers.26.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.26.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.26.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.26.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.26.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.26.self_attn.o_proj.weight"
	],
	"target": "model.layers.26.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.27.input_layernorm.weight"
	],
	"target": "model.layers.27.input_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.27.post_attention_layernorm.weight"
	],
	"target": "model.layers.27.post_attention_layernorm.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.27.shared_mlp.input_linear.weight"
	],
	"target": "model.layers.27.shared_mlp.input_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.27.shared_mlp.output_linear.weight"
	],
	"target": "model.layers.27.shared_mlp.output_linear.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.27.self_attn.o_proj.weight"
	],
	"target": "model.layers.27.self_attn.o_proj.base_layer.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.norm.weight"
	],
	"target": "model.norm.weight",
	"type": "direct"
	},
	{
	"source": [
	"lm_head.weight"
	],
	"target": "lm_head.weight",
	"type": "direct"
	},
	{
	"source": [
	"model.layers.0.self_attn.q_proj.weight",
	"model.layers.0.self_attn.k_proj.weight",
	"model.layers.0.self_attn.v_proj.weight"
	],
	"target": "model.layers.0.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.1.self_attn.q_proj.weight",
	"model.layers.1.self_attn.k_proj.weight",
	"model.layers.1.self_attn.v_proj.weight"
	],
	"target": "model.layers.1.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.2.self_attn.q_proj.weight",
	"model.layers.2.self_attn.k_proj.weight",
	"model.layers.2.self_attn.v_proj.weight"
	],
	"target": "model.layers.2.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.3.self_attn.q_proj.weight",
	"model.layers.3.self_attn.k_proj.weight",
	"model.layers.3.self_attn.v_proj.weight"
	],
	"target": "model.layers.3.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.4.self_attn.q_proj.weight",
	"model.layers.4.self_attn.k_proj.weight",
	"model.layers.4.self_attn.v_proj.weight"
	],
	"target": "model.layers.4.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.5.self_attn.q_proj.weight",
	"model.layers.5.self_attn.k_proj.weight",
	"model.layers.5.self_attn.v_proj.weight"
	],
	"target": "model.layers.5.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.6.self_attn.q_proj.weight",
	"model.layers.6.self_attn.k_proj.weight",
	"model.layers.6.self_attn.v_proj.weight"
	],
	"target": "model.layers.6.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.7.self_attn.q_proj.weight",
	"model.layers.7.self_attn.k_proj.weight",
	"model.layers.7.self_attn.v_proj.weight"
	],
	"target": "model.layers.7.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.8.self_attn.q_proj.weight",
	"model.layers.8.self_attn.k_proj.weight",
	"model.layers.8.self_attn.v_proj.weight"
	],
	"target": "model.layers.8.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.9.self_attn.q_proj.weight",
	"model.layers.9.self_attn.k_proj.weight",
	"model.layers.9.self_attn.v_proj.weight"
	],
	"target": "model.layers.9.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.10.self_attn.q_proj.weight",
	"model.layers.10.self_attn.k_proj.weight",
	"model.layers.10.self_attn.v_proj.weight"
	],
	"target": "model.layers.10.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.11.self_attn.q_proj.weight",
	"model.layers.11.self_attn.k_proj.weight",
	"model.layers.11.self_attn.v_proj.weight"
	],
	"target": "model.layers.11.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.12.self_attn.q_proj.weight",
	"model.layers.12.self_attn.k_proj.weight",
	"model.layers.12.self_attn.v_proj.weight"
	],
	"target": "model.layers.12.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.13.self_attn.q_proj.weight",
	"model.layers.13.self_attn.k_proj.weight",
	"model.layers.13.self_attn.v_proj.weight"
	],
	"target": "model.layers.13.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.14.self_attn.q_proj.weight",
	"model.layers.14.self_attn.k_proj.weight",
	"model.layers.14.self_attn.v_proj.weight"
	],
	"target": "model.layers.14.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.15.self_attn.q_proj.weight",
	"model.layers.15.self_attn.k_proj.weight",
	"model.layers.15.self_attn.v_proj.weight"
	],
	"target": "model.layers.15.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.16.self_attn.q_proj.weight",
	"model.layers.16.self_attn.k_proj.weight",
	"model.layers.16.self_attn.v_proj.weight"
	],
	"target": "model.layers.16.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.17.self_attn.q_proj.weight",
	"model.layers.17.self_attn.k_proj.weight",
	"model.layers.17.self_attn.v_proj.weight"
	],
	"target": "model.layers.17.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.18.self_attn.q_proj.weight",
	"model.layers.18.self_attn.k_proj.weight",
	"model.layers.18.self_attn.v_proj.weight"
	],
	"target": "model.layers.18.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.19.self_attn.q_proj.weight",
	"model.layers.19.self_attn.k_proj.weight",
	"model.layers.19.self_attn.v_proj.weight"
	],
	"target": "model.layers.19.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.20.self_attn.q_proj.weight",
	"model.layers.20.self_attn.k_proj.weight",
	"model.layers.20.self_attn.v_proj.weight"
	],
	"target": "model.layers.20.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.21.self_attn.q_proj.weight",
	"model.layers.21.self_attn.k_proj.weight",
	"model.layers.21.self_attn.v_proj.weight"
	],
	"target": "model.layers.21.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.22.self_attn.q_proj.weight",
	"model.layers.22.self_attn.k_proj.weight",
	"model.layers.22.self_attn.v_proj.weight"
	],
	"target": "model.layers.22.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.23.self_attn.q_proj.weight",
	"model.layers.23.self_attn.k_proj.weight",
	"model.layers.23.self_attn.v_proj.weight"
	],
	"target": "model.layers.23.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.24.self_attn.q_proj.weight",
	"model.layers.24.self_attn.k_proj.weight",
	"model.layers.24.self_attn.v_proj.weight"
	],
	"target": "model.layers.24.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.25.self_attn.q_proj.weight",
	"model.layers.25.self_attn.k_proj.weight",
	"model.layers.25.self_attn.v_proj.weight"
	],
	"target": "model.layers.25.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.26.self_attn.q_proj.weight",
	"model.layers.26.self_attn.k_proj.weight",
	"model.layers.26.self_attn.v_proj.weight"
	],
	"target": "model.layers.26.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	},
	{
	"source": [
	"model.layers.27.self_attn.q_proj.weight",
	"model.layers.27.self_attn.k_proj.weight",
	"model.layers.27.self_attn.v_proj.weight"
	],
	"target": "model.layers.27.self_attn.qkv_proj.base_layer.weight",
	"type": "fused_qkv_proj"
	}
	],
	"adapter_mapping": [
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.0.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.0.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.0.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.0.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.0.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.0.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.0.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.0.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.0.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.0.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.0.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.0.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.0.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.0.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.0.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.0.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.0.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.0.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.0.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.0.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.0.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.0.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.0.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.0.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.0.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.0.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.0.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.0.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.0.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.0.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.0.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.0.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.0.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.0.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.0.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.0.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.0.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.0.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.0.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.0.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.0.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.0.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.0.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.1.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.1.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.1.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.1.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.1.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.1.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.1.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.1.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.1.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.1.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.1.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.1.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.1.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.1.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.1.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.1.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.1.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.1.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.1.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.1.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.1.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.1.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.1.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.1.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.1.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.1.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.1.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.1.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.1.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.1.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.1.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.1.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.1.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.1.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.1.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.1.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.1.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.1.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.1.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.1.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.1.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.1.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.1.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.10.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.10.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.10.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.10.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.10.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.10.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.10.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.10.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.10.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.10.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.10.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.10.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.10.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.10.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.10.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.10.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.10.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.10.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.10.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.10.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.10.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.10.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.10.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.10.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.10.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.10.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.10.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.10.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.10.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.10.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.10.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.10.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.10.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.10.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.10.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.10.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.10.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.10.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.10.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.10.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.10.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.10.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.10.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.11.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.11.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.11.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.11.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.11.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.11.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.11.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.11.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.11.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.11.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.11.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.11.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.11.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.11.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.11.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.11.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.11.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.11.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.11.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.11.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.11.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.11.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.11.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.11.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.11.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.11.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.11.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.11.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.11.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.11.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.11.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.11.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.11.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.11.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.11.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.11.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.11.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.11.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.11.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.11.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.11.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.11.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.11.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.12.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.12.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.12.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.12.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.12.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.12.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.12.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.12.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.12.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.12.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.12.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.12.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.12.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.12.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.12.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.12.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.12.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.12.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.12.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.12.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.12.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.12.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.12.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.12.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.12.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.12.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.12.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.12.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.12.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.12.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.12.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.12.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.12.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.12.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.12.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.12.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.12.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.12.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.12.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.12.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.12.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.12.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.12.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.13.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.13.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.13.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.13.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.13.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.13.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.13.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.13.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.13.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.13.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.13.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.13.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.13.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.13.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.13.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.13.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.13.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.13.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.13.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.13.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.13.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.13.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.13.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.13.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.13.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.13.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.13.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.13.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.13.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.13.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.13.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.13.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.13.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.13.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.13.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.13.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.13.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.13.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.13.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.13.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.13.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.13.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.13.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.14.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.14.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.14.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.14.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.14.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.14.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.14.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.14.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.14.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.14.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.14.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.14.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.14.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.14.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.14.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.14.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.14.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.14.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.14.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.14.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.14.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.14.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.14.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.14.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.14.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.14.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.14.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.14.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.14.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.14.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.14.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.14.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.14.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.14.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.14.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.14.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.14.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.14.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.14.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.14.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.14.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.14.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.14.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.15.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.15.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.15.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.15.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.15.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.15.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.15.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.15.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.15.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.15.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.15.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.15.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.15.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.15.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.15.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.15.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.15.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.15.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.15.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.15.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.15.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.15.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.15.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.15.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.15.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.15.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.15.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.15.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.15.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.15.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.15.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.15.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.15.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.15.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.15.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.15.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.15.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.15.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.15.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.15.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.15.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.15.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.15.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.16.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.16.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.16.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.16.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.16.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.16.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.16.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.16.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.16.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.16.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.16.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.16.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.16.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.16.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.16.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.16.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.16.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.16.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.16.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.16.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.16.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.16.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.16.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.16.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.16.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.16.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.16.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.16.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.16.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.16.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.16.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.16.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.16.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.16.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.16.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.16.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.16.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.16.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.16.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.16.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.16.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.16.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.16.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.17.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.17.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.17.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.17.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.17.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.17.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.17.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.17.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.17.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.17.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.17.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.17.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.17.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.17.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.17.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.17.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.17.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.17.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.17.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.17.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.17.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.17.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.17.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.17.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.17.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.17.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.17.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.17.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.17.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.17.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.17.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.17.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.17.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.17.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.17.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.17.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.17.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.17.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.17.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.17.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.17.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.17.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.17.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.18.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.18.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.18.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.18.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.18.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.18.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.18.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.18.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.18.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.18.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.18.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.18.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.18.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.18.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.18.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.18.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.18.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.18.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.18.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.18.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.18.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.18.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.18.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.18.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.18.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.18.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.18.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.18.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.18.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.18.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.18.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.18.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.18.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.18.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.18.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.18.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.18.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.18.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.18.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.18.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.18.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.18.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.18.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.19.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.19.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.19.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.19.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.19.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.19.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.19.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.19.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.19.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.19.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.19.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.19.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.19.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.19.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.19.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.19.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.19.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.19.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.19.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.19.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.19.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.19.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.19.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.19.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.19.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.19.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.19.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.19.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.19.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.19.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.19.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.19.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.19.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.19.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.19.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.19.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.19.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.19.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.19.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.19.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.19.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.19.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.19.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.2.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.2.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.2.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.2.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.2.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.2.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.2.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.2.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.2.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.2.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.2.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.2.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.2.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.2.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.2.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.2.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.2.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.2.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.2.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.2.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.2.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.2.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.2.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.2.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.2.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.2.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.2.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.2.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.2.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.2.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.2.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.2.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.2.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.2.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.2.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.2.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.2.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.2.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.2.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.2.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.2.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.2.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.2.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.20.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.20.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.20.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.20.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.20.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.20.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.20.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.20.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.20.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.20.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.20.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.20.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.20.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.20.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.20.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.20.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.20.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.20.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.20.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.20.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.20.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.20.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.20.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.20.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.20.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.20.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.20.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.20.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.20.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.20.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.20.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.20.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.20.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.20.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.20.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.20.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.20.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.20.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.20.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.20.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.20.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.20.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.20.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.21.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.21.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.21.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.21.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.21.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.21.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.21.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.21.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.21.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.21.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.21.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.21.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.21.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.21.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.21.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.21.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.21.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.21.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.21.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.21.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.21.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.21.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.21.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.21.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.21.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.21.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.21.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.21.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.21.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.21.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.21.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.21.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.21.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.21.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.21.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.21.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.21.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.21.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.21.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.21.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.21.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.21.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.21.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.22.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.22.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.22.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.22.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.22.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.22.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.22.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.22.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.22.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.22.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.22.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.22.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.22.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.22.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.22.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.22.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.22.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.22.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.22.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.22.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.22.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.22.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.22.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.22.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.22.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.22.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.22.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.22.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.22.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.22.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.22.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.22.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.22.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.22.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.22.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.22.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.22.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.22.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.22.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.22.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.22.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.22.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.22.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.23.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.23.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.23.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.23.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.23.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.23.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.23.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.23.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.23.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.23.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.23.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.23.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.23.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.23.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.23.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.23.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.23.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.23.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.23.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.23.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.23.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.23.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.23.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.23.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.23.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.23.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.23.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.23.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.23.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.23.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.23.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.23.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.23.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.23.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.23.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.23.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.23.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.23.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.23.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.23.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.23.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.23.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.23.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.24.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.24.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.24.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.24.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.24.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.24.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.24.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.24.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.24.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.24.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.24.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.24.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.24.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.24.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.24.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.24.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.24.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.24.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.24.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.24.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.24.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.24.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.24.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.24.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.24.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.24.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.24.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.24.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.24.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.24.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.24.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.24.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.24.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.24.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.24.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.24.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.24.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.24.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.24.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.24.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.24.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.24.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.24.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.25.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.25.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.25.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.25.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.25.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.25.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.25.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.25.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.25.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.25.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.25.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.25.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.25.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.25.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.25.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.25.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.25.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.25.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.25.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.25.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.25.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.25.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.25.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.25.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.25.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.25.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.25.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.25.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.25.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.25.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.25.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.25.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.25.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.25.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.25.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.25.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.25.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.25.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.25.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.25.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.25.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.25.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.25.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.26.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.26.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.26.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.26.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.26.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.26.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.26.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.26.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.26.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.26.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.26.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.26.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.26.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.26.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.26.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.26.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.26.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.26.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.26.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.26.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.26.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.26.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.26.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.26.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.26.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.26.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.26.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.26.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.26.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.26.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.26.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.26.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.26.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.26.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.26.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.26.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.26.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.26.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.26.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.26.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.26.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.26.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.26.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.27.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.27.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.27.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.27.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.27.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.27.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.27.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.27.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.27.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.27.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.27.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.27.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.27.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.27.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.27.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.27.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.27.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.27.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.27.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.27.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.27.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.27.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.27.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.27.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.27.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.27.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.27.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.27.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.27.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.27.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.27.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.27.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.27.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.27.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.27.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.27.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.27.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.27.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.27.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.27.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.27.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.27.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.27.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.3.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.3.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.3.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.3.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.3.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.3.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.3.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.3.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.3.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.3.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.3.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.3.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.3.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.3.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.3.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.3.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.3.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.3.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.3.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.3.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.3.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.3.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.3.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.3.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.3.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.3.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.3.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.3.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.3.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.3.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.3.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.3.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.3.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.3.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.3.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.3.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.3.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.3.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.3.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.3.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.3.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.3.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.3.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.4.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.4.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.4.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.4.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.4.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.4.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.4.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.4.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.4.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.4.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.4.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.4.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.4.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.4.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.4.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.4.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.4.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.4.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.4.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.4.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.4.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.4.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.4.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.4.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.4.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.4.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.4.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.4.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.4.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.4.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.4.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.4.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.4.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.4.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.4.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.4.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.4.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.4.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.4.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.4.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.4.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.4.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.4.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.5.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.5.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.5.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.5.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.5.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.5.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.5.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.5.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.5.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.5.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.5.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.5.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.5.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.5.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.5.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.5.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.5.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.5.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.5.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.5.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.5.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.5.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.5.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.5.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.5.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.5.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.5.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.5.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.5.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.5.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.5.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.5.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.5.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.5.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.5.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.5.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.5.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.5.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.5.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.5.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.5.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.5.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.5.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.6.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.6.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.6.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.6.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.6.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.6.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.6.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.6.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.6.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.6.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.6.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.6.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.6.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.6.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.6.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.6.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.6.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.6.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.6.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.6.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.6.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.6.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.6.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.6.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.6.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.6.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.6.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.6.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.6.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.6.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.6.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.6.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.6.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.6.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.6.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.6.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.6.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.6.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.6.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.6.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.6.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.6.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.6.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.7.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.7.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.7.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.7.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.7.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.7.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.7.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.7.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.7.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.7.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.7.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.7.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.7.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.7.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.7.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.7.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.7.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.7.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.7.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.7.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.7.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.7.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.7.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.7.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.7.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.7.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.7.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.7.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.7.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.7.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.7.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.7.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.7.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.7.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.7.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.7.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.7.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.7.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.7.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.7.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.7.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.7.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.7.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.8.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.8.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.8.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.8.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.8.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.8.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.8.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.8.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.8.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.8.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.8.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.8.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.8.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.8.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.8.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.8.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.8.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.8.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.8.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.8.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.8.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.8.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.8.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.8.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.8.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.8.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.8.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.8.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.8.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.8.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.8.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.8.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.8.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.8.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.8.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.8.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.8.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.8.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.8.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.8.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.8.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.8.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.8.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.self_attn.o_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.9.self_attn.o_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.9.self_attn.o_proj.lora_A.weight"
	],
	"target": "model.layers.9.self_attn.o_proj.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.self_attn.o_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.9.self_attn.o_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.9.self_attn.o_proj.lora_B.weight"
	],
	"target": "model.layers.9.self_attn.o_proj.lora_B",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.self_attn.q_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.9.self_attn.q_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.9.self_attn.q_proj.lora_A.weight"
	],
	"target": "model.layers.9.self_attn.qkv_proj.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.self_attn.k_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.9.self_attn.k_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.9.self_attn.k_proj.lora_A.weight"
	],
	"target": "model.layers.9.self_attn.qkv_proj.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.self_attn.v_proj.lora_A.weight",
	"adapter_1::base_model.model.model.layers.9.self_attn.v_proj.lora_A.weight",
	"adapter_2::base_model.model.model.layers.9.self_attn.v_proj.lora_A.weight"
	],
	"target": "model.layers.9.self_attn.qkv_proj.lora_A_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.self_attn.q_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.9.self_attn.q_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.9.self_attn.q_proj.lora_B.weight"
	],
	"target": "model.layers.9.self_attn.qkv_proj.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.self_attn.k_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.9.self_attn.k_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.9.self_attn.k_proj.lora_B.weight"
	],
	"target": "model.layers.9.self_attn.qkv_proj.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.self_attn.v_proj.lora_B.weight",
	"adapter_1::base_model.model.model.layers.9.self_attn.v_proj.lora_B.weight",
	"adapter_2::base_model.model.model.layers.9.self_attn.v_proj.lora_B.weight"
	],
	"target": "model.layers.9.self_attn.qkv_proj.lora_B_slices.2",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.9.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.9.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.9.shared_mlp.input_linear.lora_A_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.shared_mlp.input_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.9.shared_mlp.input_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.9.shared_mlp.input_linear.lora_A.weight"
	],
	"target": "model.layers.9.shared_mlp.input_linear.lora_A_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.9.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.9.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.9.shared_mlp.input_linear.lora_B_slices.0",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.shared_mlp.input_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.9.shared_mlp.input_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.9.shared_mlp.input_linear.lora_B.weight"
	],
	"target": "model.layers.9.shared_mlp.input_linear.lora_B_slices.1",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.shared_mlp.output_linear.lora_A.weight",
	"adapter_1::base_model.model.model.layers.9.shared_mlp.output_linear.lora_A.weight",
	"adapter_2::base_model.model.model.layers.9.shared_mlp.output_linear.lora_A.weight"
	],
	"target": "model.layers.9.shared_mlp.output_linear.lora_A",
	"type": "adapter_stacked"
	},
	{
	"source": [
	"adapter_0::base_model.model.model.layers.9.shared_mlp.output_linear.lora_B.weight",
	"adapter_1::base_model.model.model.layers.9.shared_mlp.output_linear.lora_B.weight",
	"adapter_2::base_model.model.model.layers.9.shared_mlp.output_linear.lora_B.weight"
	],
	"target": "model.layers.9.shared_mlp.output_linear.lora_B",
	"type": "adapter_stacked"
	}
	]
	}