Instructions to use lamekyemane09/tigrinya-human-model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lamekyemane09/tigrinya-human-model with Transformers:

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

pipe = pipeline("text-generation", model="lamekyemane09/tigrinya-human-model")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("lamekyemane09/tigrinya-human-model")
model = AutoModelForCausalLM.from_pretrained("lamekyemane09/tigrinya-human-model")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use lamekyemane09/tigrinya-human-model with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lamekyemane09/tigrinya-human-model"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lamekyemane09/tigrinya-human-model",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/lamekyemane09/tigrinya-human-model

SGLang

How to use lamekyemane09/tigrinya-human-model 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 "lamekyemane09/tigrinya-human-model" \
    --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": "lamekyemane09/tigrinya-human-model",
		"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 "lamekyemane09/tigrinya-human-model" \
        --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": "lamekyemane09/tigrinya-human-model",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use lamekyemane09/tigrinya-human-model with Docker Model Runner:
```
docker model run hf.co/lamekyemane09/tigrinya-human-model
```

tigrinya-human-model

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

Loss: 6.2045

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: 0.0003
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 200
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
8.0394	0.1142	100	7.7904
7.6959	0.2284	200	7.6207
7.5973	0.3426	300	7.4871
7.4921	0.4568	400	7.3849
7.4058	0.5709	500	7.3094
7.2915	0.6851	600	7.1744
7.1601	0.7993	700	7.0875
7.0527	0.9135	800	6.9761
6.9395	1.0274	900	6.9111
6.8182	1.1416	1000	6.8366
6.7701	1.2558	1100	6.7599
6.7175	1.3700	1200	6.7156
6.6768	1.4842	1300	6.6487
6.6053	1.5983	1400	6.6034
6.5911	1.7125	1500	6.5556
6.5469	1.8267	1600	6.5132
6.4954	1.9409	1700	6.4833
6.2197	2.0548	1800	6.4587
6.2399	2.1690	1900	6.4149
6.1844	2.2832	2000	6.3713
6.1713	2.3974	2100	6.3447
6.1337	2.5116	2200	6.3205
6.0964	2.6257	2300	6.2838
6.1045	2.7399	2400	6.2665
6.0403	2.8541	2500	6.2431
6.0334	2.9683	2600	6.2070
5.6972	3.0822	2700	6.2313
5.7433	3.1964	2800	6.2223
5.6886	3.3106	2900	6.2213
5.7266	3.4248	3000	6.1778
5.6925	3.5390	3100	6.1677
5.7041	3.6532	3200	6.1701
5.7159	3.7673	3300	6.1559
5.6899	3.8815	3400	6.1493
5.7043	3.9957	3500	6.1856
5.6214	4.1096	3600	6.2702
5.7020	4.2238	3700	6.2724
5.6668	4.3380	3800	6.2092
5.7011	4.4522	3900	6.2258
5.7572	4.5664	4000	6.2257

Framework versions

Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 4.8.5
Tokenizers 0.22.2

Downloads last month: 852

Safetensors

Model size

0.3B params

Tensor type

F32