Instructions to use HachiML/myBit-Llama2-jp-127M-test-7 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HachiML/myBit-Llama2-jp-127M-test-7 with Transformers:

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

pipe = pipeline("text-generation", model="HachiML/myBit-Llama2-jp-127M-test-7")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("HachiML/myBit-Llama2-jp-127M-test-7")
model = AutoModelForCausalLM.from_pretrained("HachiML/myBit-Llama2-jp-127M-test-7", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use HachiML/myBit-Llama2-jp-127M-test-7 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "HachiML/myBit-Llama2-jp-127M-test-7"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "HachiML/myBit-Llama2-jp-127M-test-7",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/HachiML/myBit-Llama2-jp-127M-test-7

SGLang

How to use HachiML/myBit-Llama2-jp-127M-test-7 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/myBit-Llama2-jp-127M-test-7" \
    --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/myBit-Llama2-jp-127M-test-7",
		"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/myBit-Llama2-jp-127M-test-7" \
        --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/myBit-Llama2-jp-127M-test-7",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use HachiML/myBit-Llama2-jp-127M-test-7 with Docker Model Runner:
```
docker model run hf.co/HachiML/myBit-Llama2-jp-127M-test-7
```

myBit-Llama2-jp-127M-7

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: 10.6539

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: 0.00024
train_batch_size: 96
eval_batch_size: 96
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: polynomial
lr_scheduler_warmup_steps: 250
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
9.0536	0.04	100	7.4802
6.8962	0.07	200	6.5875
6.3685	0.11	300	6.1149
5.8698	0.15	400	5.6208
5.6334	0.18	500	6.1096
8.8705	0.22	600	10.3915
10.5174	0.26	700	10.5752
10.5929	0.29	800	10.6066
10.6128	0.33	900	10.6187
10.6218	0.37	1000	10.6255
10.6274	0.4	1100	10.6302
10.6312	0.44	1200	10.6335
10.6343	0.48	1300	10.6363
10.6369	0.51	1400	10.6384
10.6391	0.55	1500	10.6404
10.6408	0.59	1600	10.6422
10.6426	0.62	1700	10.6438
10.6441	0.66	1800	10.6451
10.6454	0.7	1900	10.6464
10.6467	0.73	2000	10.6477
10.6479	0.77	2100	10.6486
10.649	0.81	2200	10.6496
10.6499	0.84	2300	10.6506
10.6508	0.88	2400	10.6515
10.6516	0.92	2500	10.6522
10.6524	0.95	2600	10.6531
10.6534	0.99	2700	10.6539

Framework versions

Transformers 4.38.2
Pytorch 2.1.0+cu121
Datasets 2.18.0
Tokenizers 0.15.2

Downloads last month: 8

Safetensors

Model size

0.1B params

Tensor type

F32

Model tree for HachiML/myBit-Llama2-jp-127M-test-7

Base model

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

Finetuned

(575)

this model