Instructions to use shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b with Transformers:

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

pipe = pipeline("text-generation", model="shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b")
model = AutoModelForCausalLM.from_pretrained("shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b

SGLang

How to use shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b 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 "shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b" \
    --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": "shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b",
		"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 "shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b" \
        --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": "shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b with Docker Model Runner:
```
docker model run hf.co/shisa-ai/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

See axolotl config

axolotl version: 0.6.0

# train w/ shisa-ai/shisa-v1-athenev2-reannotated-filtered

base_model: allenai/Llama-3.1-Tulu-3-8B
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: false
strict: false

# User Liger
plugins:
  - axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_fused_linear_cross_entropy: true

chat_template: llama3
datasets:
  - path: shisa-ai/shisa-v1-athenev2-reannotated-filtered
    # type: sharegpt deprecated
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
dataset_prepared_path: last_run_prepared
val_set_size: 0.05
output_dir: ./outputs/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b

sequence_len: 8192
sample_packing: true
pad_to_sequence_len: true

# marginal difference
neftune_noise_alpha: 5

use_wandb: true
wandb_project: shisa-v2
wandb_entity: augmxnt
wandb_name: ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b

gradient_accumulation_steps: 2
micro_batch_size: 4
num_epochs: 3
optimizer: paged_adamw_8bit
lr_scheduler: linear
learning_rate: 8e-6

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false

gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_ratio: 0.05
evals_per_epoch: 2
eval_table_size:
saves_per_epoch: 0
save_total_limit: 1 # Only store a single checkpoint
debug:
deepspeed: zero3_bf16.json
weight_decay: 0.00
fsdp:
fsdp_config:
special_tokens:
  pad_token: <|end_of_text|>

outputs/ablation-28-rafathenev2.tulu-shisa-v2-tulu3-8b

This model is a fine-tuned version of allenai/Llama-3.1-Tulu-3-8B on the shisa-ai/shisa-v1-athenev2-reannotated-filtered dataset. It achieves the following results on the evaluation set:

Loss: 0.4476

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 8e-06
train_batch_size: 4
eval_batch_size: 4
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 2
total_train_batch_size: 64
total_eval_batch_size: 32
optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 25
num_epochs: 3.0