Instructions to use zaursamedov1/Llama2-ft-qa with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zaursamedov1/Llama2-ft-qa with Transformers:

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

pipe = pipeline("text-generation", model="zaursamedov1/Llama2-ft-qa")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("zaursamedov1/Llama2-ft-qa")
model = AutoModelForCausalLM.from_pretrained("zaursamedov1/Llama2-ft-qa")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use zaursamedov1/Llama2-ft-qa with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zaursamedov1/Llama2-ft-qa"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zaursamedov1/Llama2-ft-qa",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/zaursamedov1/Llama2-ft-qa

SGLang

How to use zaursamedov1/Llama2-ft-qa 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 "zaursamedov1/Llama2-ft-qa" \
    --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": "zaursamedov1/Llama2-ft-qa",
		"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 "zaursamedov1/Llama2-ft-qa" \
        --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": "zaursamedov1/Llama2-ft-qa",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use zaursamedov1/Llama2-ft-qa with Docker Model Runner:
```
docker model run hf.co/zaursamedov1/Llama2-ft-qa
```

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Usage of this model:

I'm glad to share with you my exciting journey of fine-tuning Llama 2 for Named Entity Recognition (NER),particularly on a customer service dataset. NER is a fascinating natural language processing task that involves identifying and classifying entities like names of people, organizations, locations, and other important terms within a given text.

The customer service dataset I used was carefully curated and annotated with a wide range of service-related entities, such as specific types of services, service providers, service locations, and other related terms. The data was diverse and representative of the actual domain it aimed to address. (I will re-upload the dataset with more sample in it to here zaursamedov1/customer-service-ner)

To get more closer look at to the model read this colab notebook

(Coming soon...)

library_name: peft

Training procedure

The following bitsandbytes quantization config was used during training:

load_in_8bit: False
load_in_4bit: True
llm_int8_threshold: 6.0
llm_int8_skip_modules: None
llm_int8_enable_fp32_cpu_offload: False
llm_int8_has_fp16_weight: False
bnb_4bit_quant_type: nf4
bnb_4bit_use_double_quant: False
bnb_4bit_compute_dtype: float16

Framework versions

PEFT 0.5.0.dev0

Downloads last month: 15