Instructions to use Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3 with Transformers:

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

pipe = pipeline("text-generation", model="Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3")
model = AutoModelForMultimodalLM.from_pretrained("Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3

SGLang

How to use Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3 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 "Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3" \
    --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": "Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3",
		"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 "Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3" \
        --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": "Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3 with Docker Model Runner:
```
docker model run hf.co/Minbyul/meditron-7b-dpo-full-wo-healthsearch_qa-ep3
```

meditron-7b-dpo-full-wo-healthsearch_qa-ep3 / README.md

Minbyul

End of training

5ee2b17 verified about 2 years ago

preview code

raw

history blame contribute delete

1.73 kB

metadata

license: llama2
base_model: epfl-llm/meditron-7b
tags:
  - alignment-handbook
  - trl
  - dpo
  - generated_from_trainer
  - trl
  - dpo
  - generated_from_trainer
datasets:
  - HuggingFaceH4/ultrafeedback_binarized
model-index:
  - name: meditron-7b-dpo-full-wo-healthsearch_qa-ep3
    results: []

meditron-7b-dpo-full-wo-healthsearch_qa-ep3

This model is a fine-tuned version of epfl-llm/meditron-7b on the HuggingFaceH4/ultrafeedback_binarized dataset. It achieves the following results on the evaluation set:

Loss: 0.6786
Rewards/chosen: -0.0040
Rewards/rejected: -0.0362
Rewards/accuracies: 0.6245
Rewards/margins: 0.0322
Logps/rejected: -1242.8666
Logps/chosen: -1081.9209
Logits/rejected: -0.7759
Logits/chosen: -0.8007

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: 5e-07
train_batch_size: 8
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 2
total_train_batch_size: 64
total_eval_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 1

Training results

Framework versions

Transformers 4.39.0.dev0
Pytorch 2.1.2
Datasets 2.14.6
Tokenizers 0.15.2