Instructions to use tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
model = PeftModel.from_pretrained(base_model, "tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps")

Transformers

How to use tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps with Transformers:

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

pipe = pipeline("text-generation", model="tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps

SGLang

How to use tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps 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 "tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps with Docker Model Runner:
```
docker model run hf.co/tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps
```

[8th March 2026 Update]: Upload a new model (overriding the original one) without adapter on lm_head. Lora adapter on lm_head invokes a subtle issue on Peft that leads to device mismatch when using multi-gpus.

Model Card for Model ID

This model replicates LatentQA activation decoder for Llama-3.1-8B-Instruct. The original release of LatentQA decoder is Llama-3-8B-Instruct-based. This model has the same explantion results on the "promote_veganism" example documented in the repo of original work.