Instructions to use kalpeshk2011/instruct-llama-7b-wdiff with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kalpeshk2011/instruct-llama-7b-wdiff with Transformers:

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

pipe = pipeline("text-generation", model="kalpeshk2011/instruct-llama-7b-wdiff")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kalpeshk2011/instruct-llama-7b-wdiff")
model = AutoModelForCausalLM.from_pretrained("kalpeshk2011/instruct-llama-7b-wdiff")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use kalpeshk2011/instruct-llama-7b-wdiff with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kalpeshk2011/instruct-llama-7b-wdiff"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kalpeshk2011/instruct-llama-7b-wdiff",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/kalpeshk2011/instruct-llama-7b-wdiff

SGLang

How to use kalpeshk2011/instruct-llama-7b-wdiff 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 "kalpeshk2011/instruct-llama-7b-wdiff" \
    --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": "kalpeshk2011/instruct-llama-7b-wdiff",
		"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 "kalpeshk2011/instruct-llama-7b-wdiff" \
        --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": "kalpeshk2011/instruct-llama-7b-wdiff",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use kalpeshk2011/instruct-llama-7b-wdiff with Docker Model Runner:
```
docker model run hf.co/kalpeshk2011/instruct-llama-7b-wdiff
```

This is the HuggingFace model release of the instruction tuned LLAMA-7B model used in our paper FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation.

Please refer to the README for instructions on how to setup the model (link).

Credits to Yizhong Wang for originally training this model.

Downloads last month: 18

Paper for kalpeshk2011/instruct-llama-7b-wdiff

FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation

Paper • 2305.14251 • Published May 23, 2023 • 2