Instructions to use fbellame/mistral-7b-json-quizz-fine-tuned with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use fbellame/mistral-7b-json-quizz-fine-tuned with Transformers:

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

pipe = pipeline("text-generation", model="fbellame/mistral-7b-json-quizz-fine-tuned")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("fbellame/mistral-7b-json-quizz-fine-tuned")
model = AutoModelForMultimodalLM.from_pretrained("fbellame/mistral-7b-json-quizz-fine-tuned")
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 fbellame/mistral-7b-json-quizz-fine-tuned with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "fbellame/mistral-7b-json-quizz-fine-tuned"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fbellame/mistral-7b-json-quizz-fine-tuned",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/fbellame/mistral-7b-json-quizz-fine-tuned

SGLang

How to use fbellame/mistral-7b-json-quizz-fine-tuned 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 "fbellame/mistral-7b-json-quizz-fine-tuned" \
    --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": "fbellame/mistral-7b-json-quizz-fine-tuned",
		"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 "fbellame/mistral-7b-json-quizz-fine-tuned" \
        --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": "fbellame/mistral-7b-json-quizz-fine-tuned",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use fbellame/mistral-7b-json-quizz-fine-tuned with Docker Model Runner:
```
docker model run hf.co/fbellame/mistral-7b-json-quizz-fine-tuned
```

mistral-7b-json-quizz-fine-tuned

File size: 2,918 Bytes

edec775

architecture:
    backbone_dtype: int4
    force_embedding_gradients: false
    gradient_checkpointing: true
    intermediate_dropout: 0.0
    pretrained: true
    pretrained_weights: ''
augmentation:
    neftune_noise_alpha: 0.0
    random_parent_probability: 0.0
    skip_parent_probability: 0.0
    token_mask_probability: 0.0
dataset:
    add_eos_token_to_answer: true
    add_eos_token_to_prompt: false
    add_eos_token_to_system: true
    answer_column: output
    chatbot_author: H2O.ai
    chatbot_name: h2oGPT
    data_sample: 1.0
    data_sample_choice:
    - Train
    - Validation
    limit_chained_samples: false
    mask_prompt_labels: true
    parent_id_column: None
    personalize: false
    prompt_column:
    - instruction
    system_column: None
    text_answer_separator: ''
    text_prompt_start: ''
    text_system_start: <|system|>
    train_dataframe: /mnt/ssd2/h2o-llmstudio/data/user/train_data.2/train_data.parquet
    validation_dataframe: /mnt/ssd2/h2o-llmstudio/data/user/train_data.2/validate_data.parquet
    validation_size: 0.01
    validation_strategy: custom
environment:
    compile_model: false
    deepspeed_reduce_bucket_size: 1000000
    deepspeed_stage3_param_persistence_threshold: 1000000
    deepspeed_stage3_prefetch_bucket_size: 1000000
    find_unused_parameters: false
    gpus:
    - '0'
    huggingface_branch: main
    mixed_precision: true
    number_of_workers: 8
    seed: -1
    trust_remote_code: true
    use_deepspeed: false
experiment_name: mistral_7b_json_quizz.fine_tuned
llm_backbone: mistralai/Mistral-7B-Instruct-v0.2
logging:
    logger: None
    neptune_project: ''
output_directory: /mnt/ssd2/h2o-llmstudio/output/user/mistral_7b_json_quizz.fine_tuned/
prediction:
    batch_size_inference: 0
    do_sample: false
    max_length_inference: 256
    metric: GPT
    metric_gpt_model: gpt-3.5-turbo-0613
    metric_gpt_template: general
    min_length_inference: 2
    num_beams: 1
    num_history: 4
    repetition_penalty: 1.0
    stop_tokens: ''
    temperature: 0.0
    top_k: 0
    top_p: 1.0
problem_type: text_causal_language_modeling
tokenizer:
    add_prefix_space: false
    add_prompt_answer_tokens: false
    max_length: 1024
    max_length_answer: 512
    max_length_prompt: 512
    padding_quantile: 1.0
    use_fast: true
training:
    batch_size: 1
    differential_learning_rate: 1.0e-05
    differential_learning_rate_layers: []
    drop_last_batch: true
    epochs: 2
    evaluate_before_training: true
    evaluation_epochs: 1.0
    grad_accumulation: 1
    gradient_clip: 0.0
    learning_rate: 0.0001
    lora: true
    lora_alpha: 16
    lora_dropout: 0.05
    lora_r: 4
    lora_target_modules: ''
    loss_function: TokenAveragedCrossEntropy
    optimizer: AdamW
    save_best_checkpoint: false
    schedule: Cosine
    train_validation_data: false
    use_flash_attention_2: true
    warmup_epochs: 0.0
    weight_decay: 0.0