Instructions to use tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO with Transformers:

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

pipe = pipeline("text-generation", model="tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO")
model = AutoModelForCausalLM.from_pretrained("tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO", device_map="auto")
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 tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO

SGLang

How to use tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO 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 "tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO" \
    --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": "tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO",
		"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 "tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO" \
        --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": "tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO with Docker Model Runner:
```
docker model run hf.co/tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO
```

Na_M2_300steps_1e8rate_01beta_cSFTDPO

This model is a fine-tuned version of tsavage68/Na_M2_1000steps_1e7_SFT on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.6653
Rewards/chosen: 0.0172
Rewards/rejected: -0.0396
Rewards/accuracies: 0.9100
Rewards/margins: 0.0568
Logps/rejected: -80.3192
Logps/chosen: -47.9599
Logits/rejected: -2.5356
Logits/chosen: -2.5482

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: 1e-08
train_batch_size: 2
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 4
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 100
training_steps: 300

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.6929	0.2667	50	0.6931	-0.0010	-0.0014	0.5700	0.0003	-79.9371	-48.1427	-2.5355	-2.5481
0.6881	0.5333	100	0.6832	0.0062	-0.0142	0.6900	0.0204	-80.0656	-48.0704	-2.5357	-2.5482
0.67	0.8	150	0.6690	0.0151	-0.0343	0.9000	0.0494	-80.2661	-47.9813	-2.5359	-2.5485
0.6726	1.0667	200	0.6633	0.0167	-0.0444	0.9500	0.0611	-80.3677	-47.9656	-2.5356	-2.5482
0.6615	1.3333	250	0.6653	0.0172	-0.0396	0.9100	0.0568	-80.3192	-47.9599	-2.5356	-2.5482
0.667	1.6	300	0.6653	0.0172	-0.0396	0.9100	0.0568	-80.3192	-47.9599	-2.5356	-2.5482

Framework versions

Transformers 4.44.2
Pytorch 2.4.0+cu121
Datasets 2.21.0
Tokenizers 0.19.1

Downloads last month: 6

Safetensors

Model size

7B params

Tensor type

F16

Model tree for tsavage68/Na_M2_300steps_1e8rate_01beta_cSFTDPO

Base model

mistralai/Mistral-7B-Instruct-v0.2

Finetuned

tsavage68/Na_M2_1000steps_1e7_SFT

Finetuned

(11)

this model