Instructions to use dty1aaa/codellama-7b-instruct-hf-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dty1aaa/codellama-7b-instruct-hf-sft with Transformers:

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

pipe = pipeline("text-generation", model="dty1aaa/codellama-7b-instruct-hf-sft")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("dty1aaa/codellama-7b-instruct-hf-sft")
model = AutoModelForCausalLM.from_pretrained("dty1aaa/codellama-7b-instruct-hf-sft")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use dty1aaa/codellama-7b-instruct-hf-sft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dty1aaa/codellama-7b-instruct-hf-sft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dty1aaa/codellama-7b-instruct-hf-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/dty1aaa/codellama-7b-instruct-hf-sft

SGLang

How to use dty1aaa/codellama-7b-instruct-hf-sft 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 "dty1aaa/codellama-7b-instruct-hf-sft" \
    --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": "dty1aaa/codellama-7b-instruct-hf-sft",
		"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 "dty1aaa/codellama-7b-instruct-hf-sft" \
        --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": "dty1aaa/codellama-7b-instruct-hf-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use dty1aaa/codellama-7b-instruct-hf-sft with Docker Model Runner:
```
docker model run hf.co/dty1aaa/codellama-7b-instruct-hf-sft
```

codellama-7b-instruct-hf-sft

File size: 2,123 Bytes

---
library_name: transformers
license: other
base_model: codellama/CodeLlama-7b-Instruct-hf
tags:
- llama-factory
- full
- generated_from_trainer
model-index:
- name: final_sft_CodeLlama-7b-Instruct-hf
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# final_sft_CodeLlama-7b-Instruct-hf

This model is a fine-tuned version of [codellama/CodeLlama-7b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf) on the efficoder dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3331

## Model description

train on EFFIINSTRUCT_python 

## Intended uses & limitations

generate effective code

## Training and evaluation data

train on EFFIINSTRUCT_python 

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 8
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- total_eval_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 4.0

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.2475        | 0.3401 | 50   | 0.2433          |
| 0.2425        | 0.6803 | 100  | 0.2314          |
| 0.2147        | 1.0204 | 150  | 0.2318          |
| 0.1702        | 1.3605 | 200  | 0.2378          |
| 0.1758        | 1.7007 | 250  | 0.2396          |
| 0.1438        | 2.0408 | 300  | 0.2643          |
| 0.0989        | 2.3810 | 350  | 0.2823          |
| 0.0991        | 2.7211 | 400  | 0.2799          |
| 0.065         | 3.0612 | 450  | 0.3123          |
| 0.0601        | 3.4014 | 500  | 0.3348          |
| 0.0645        | 3.7415 | 550  | 0.3324          |


### Framework versions

- Transformers 4.44.2
- Pytorch 2.2.2+cu121
- Datasets 4.8.4
- Tokenizers 0.19.1