Instructions to use harborwater/open-llama-3b-everythingLM-2048 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use harborwater/open-llama-3b-everythingLM-2048 with Transformers:

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

pipe = pipeline("text-generation", model="harborwater/open-llama-3b-everythingLM-2048")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("harborwater/open-llama-3b-everythingLM-2048")
model = AutoModelForCausalLM.from_pretrained("harborwater/open-llama-3b-everythingLM-2048", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use harborwater/open-llama-3b-everythingLM-2048 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "harborwater/open-llama-3b-everythingLM-2048"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "harborwater/open-llama-3b-everythingLM-2048",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/harborwater/open-llama-3b-everythingLM-2048

SGLang

How to use harborwater/open-llama-3b-everythingLM-2048 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 "harborwater/open-llama-3b-everythingLM-2048" \
    --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": "harborwater/open-llama-3b-everythingLM-2048",
		"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 "harborwater/open-llama-3b-everythingLM-2048" \
        --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": "harborwater/open-llama-3b-everythingLM-2048",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use harborwater/open-llama-3b-everythingLM-2048 with Docker Model Runner:
```
docker model run hf.co/harborwater/open-llama-3b-everythingLM-2048
```

Trained on 2 epochs on the EverythingLM-data-V3 dataset.

This model uses the alpaca prompt format:

Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.

### Instruction:
Instruction

### Input:
Input

### Response:

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	40.62
AI2 Reasoning Challenge (25-Shot)	42.75
HellaSwag (10-Shot)	71.72
MMLU (5-Shot)	27.16
TruthfulQA (0-shot)	34.26
Winogrande (5-shot)	66.30
GSM8k (5-shot)	1.52

Downloads last month: 88

Safetensors

Model size

3B params

Tensor type

F32

F16

Dataset used to train harborwater/open-llama-3b-everythingLM-2048

Evaluation results

normalized accuracy on AI2 Reasoning Challenge (25-Shot)
test set Open LLM Leaderboard

42.750
normalized accuracy on HellaSwag (10-Shot)
validation set Open LLM Leaderboard

71.720
accuracy on MMLU (5-Shot)
test set Open LLM Leaderboard

27.160
mc2 on TruthfulQA (0-shot)
validation set Open LLM Leaderboard

34.260
accuracy on Winogrande (5-shot)
validation set Open LLM Leaderboard

66.300
accuracy on GSM8k (5-shot)
test set Open LLM Leaderboard

1.520