Instructions to use HachiML/Bit-Llama2-jp-123M-test-1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HachiML/Bit-Llama2-jp-123M-test-1 with Transformers:

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

pipe = pipeline("text-generation", model="HachiML/Bit-Llama2-jp-123M-test-1")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("HachiML/Bit-Llama2-jp-123M-test-1")
model = AutoModelForCausalLM.from_pretrained("HachiML/Bit-Llama2-jp-123M-test-1")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use HachiML/Bit-Llama2-jp-123M-test-1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "HachiML/Bit-Llama2-jp-123M-test-1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "HachiML/Bit-Llama2-jp-123M-test-1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/HachiML/Bit-Llama2-jp-123M-test-1

SGLang

How to use HachiML/Bit-Llama2-jp-123M-test-1 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 "HachiML/Bit-Llama2-jp-123M-test-1" \
    --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": "HachiML/Bit-Llama2-jp-123M-test-1",
		"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 "HachiML/Bit-Llama2-jp-123M-test-1" \
        --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": "HachiML/Bit-Llama2-jp-123M-test-1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use HachiML/Bit-Llama2-jp-123M-test-1 with Docker Model Runner:
```
docker model run hf.co/HachiML/Bit-Llama2-jp-123M-test-1
```

Bit-Llama2-jp-123M-test-1

This model is a fine-tuned version of TinyLlama/TinyLlama-1.1B-Chat-v1.0 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 4.8881

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0005
train_batch_size: 156
eval_batch_size: 156
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 200
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
81.1838	0.06	100	36.2006
19.2038	0.12	200	14.9311
11.2975	0.18	300	7.1671
6.8849	0.24	400	7.7815
6.636	0.3	500	5.9966
5.8384	0.36	600	5.6706
5.5927	0.42	700	5.4943
5.3739	0.47	800	5.4396
5.2842	0.53	900	5.2687
5.1623	0.59	1000	5.1615
5.1032	0.65	1100	5.2671
5.0324	0.71	1200	5.0162
4.9742	0.77	1300	4.9700
4.9145	0.83	1400	4.9156
4.8702	0.89	1500	4.9086
4.8465	0.95	1600	4.8881

Framework versions

Transformers 4.36.2
Pytorch 2.2.0+cu121
Datasets 2.18.0
Tokenizers 0.15.2

Downloads last month: 4

Safetensors

Model size

0.1B params

Tensor type

F32

Model tree for HachiML/Bit-Llama2-jp-123M-test-1

Base model

TinyLlama/TinyLlama-1.1B-Chat-v1.0

Finetuned

(554)

this model