Instructions to use michaelbzhu/test-7.6B-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use michaelbzhu/test-7.6B-base with Transformers:

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

pipe = pipeline("text-generation", model="michaelbzhu/test-7.6B-base", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("michaelbzhu/test-7.6B-base", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use michaelbzhu/test-7.6B-base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "michaelbzhu/test-7.6B-base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "michaelbzhu/test-7.6B-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/michaelbzhu/test-7.6B-base

SGLang

How to use michaelbzhu/test-7.6B-base 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 "michaelbzhu/test-7.6B-base" \
    --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": "michaelbzhu/test-7.6B-base",
		"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 "michaelbzhu/test-7.6B-base" \
        --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": "michaelbzhu/test-7.6B-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use michaelbzhu/test-7.6B-base with Docker Model Runner:
```
docker model run hf.co/michaelbzhu/test-7.6B-base
```

trained on 12,312,444,928 tokens from the kjj0/fineweb100B-gpt2 dataset

$ lm_eval --model hf \
    --model_args pretrained=michaelbzhu/test-7.6B-base,trust_remote_code=True \
    --tasks mmlu_college_medicine,hellaswag,lambada_openai,arc_easy,winogrande,arc_challenge,openbookqa \
    --device cuda:0 \
    --batch_size 16

|     Tasks      |Version|Filter|n-shot|  Metric  |   | Value |   |Stderr|
|----------------|------:|------|-----:|----------|---|------:|---|-----:|
|arc_challenge   |      1|none  |     0|acc       |↑  | 0.2295|±  |0.0123|
|                |       |none  |     0|acc_norm  |↑  | 0.2628|±  |0.0129|
|arc_easy        |      1|none  |     0|acc       |↑  | 0.5358|±  |0.0102|
|                |       |none  |     0|acc_norm  |↑  | 0.4663|±  |0.0102|
|hellaswag       |      1|none  |     0|acc       |↑  | 0.3788|±  |0.0048|
|                |       |none  |     0|acc_norm  |↑  | 0.4801|±  |0.0050|
|lambada_openai  |      1|none  |     0|acc       |↑  | 0.4527|±  |0.0069|
|                |       |none  |     0|perplexity|↓  |14.3601|±  |0.4468|
|college_medicine|      1|none  |     0|acc       |↑  | 0.2254|±  |0.0319|
|openbookqa      |      1|none  |     0|acc       |↑  | 0.1920|±  |0.0176|
|                |       |none  |     0|acc_norm  |↑  | 0.3020|±  |0.0206|
|winogrande      |      1|none  |     0|acc       |↑  | 0.5107|±  |0.0140|

Downloads last month: 1

Safetensors

Model size

8B params

Tensor type

F32

michaelbzhu
/

test-7.6B-base

Dataset used to train michaelbzhu/test-7.6B-base