Instructions to use yodayo-ai/nephra_v1.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yodayo-ai/nephra_v1.0 with Transformers:

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

pipe = pipeline("text-generation", model="yodayo-ai/nephra_v1.0")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("yodayo-ai/nephra_v1.0")
model = AutoModelForCausalLM.from_pretrained("yodayo-ai/nephra_v1.0", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use yodayo-ai/nephra_v1.0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yodayo-ai/nephra_v1.0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yodayo-ai/nephra_v1.0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/yodayo-ai/nephra_v1.0

SGLang

How to use yodayo-ai/nephra_v1.0 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 "yodayo-ai/nephra_v1.0" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yodayo-ai/nephra_v1.0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "yodayo-ai/nephra_v1.0" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yodayo-ai/nephra_v1.0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use yodayo-ai/nephra_v1.0 with Docker Model Runner:
```
docker model run hf.co/yodayo-ai/nephra_v1.0
```

akhazr commited on Jun 19, 2024

Commit

1c9b278

verified ·

1 Parent(s): a279423

Update README.md

Browse files

Files changed (1) hide show

README.md +0 -25

README.md CHANGED Viewed

@@ -69,31 +69,6 @@ Repetition Penalty: 1.1
 Custom Stopping Strings: "\n{{user}}", "<" , "```" , -> Has occasional broken generations.
 ```
-## Training
-These are the key hyperparameters used during training:
-| Hyperparameters                       | Finetuning              |
-|-------------------------------|----------------------------|
-| **Hardware**                  | 4x Nvidia L40 48GB        |
-| **Batch Size**                | 4x 2                         |
-| **Gradient Accumulation Steps** | 4x 3                        |
-| **LoRA Rank**              | 32                      |
-| **LoRA Alpha**              | 64                       |
-| **LoRA Dropout**              | 0.04                       |
-| **Seq_Length**              | 8192                       |
-| **LoRA Target Layers**              | All Linear Layers                      |
-| **Epochs**                    | 2                         |
-| **Max Learning Rate**        | 2e-4                       |
-| **Min Learning Rate** | 4e-5                   |
-| **Optimizer**                 | adamw_bnb_8bit                 |
-| **Optimizer Args**            | Warmup: True | Steps: 20
-| **Scheduler**                 | cosine_with_min_lr      |
-| **Warmup Steps**              | 4%                       |
 ## License
 Nephra v1 falls under [META LLAMA 3 COMMUNITY LICENSE AGREEMENT](https://huggingface.co/meta-llama/Meta-Llama-3-8B/blob/main/LICENSE).
-to

 Custom Stopping Strings: "\n{{user}}", "<" , "```" , -> Has occasional broken generations.
 ```
 ## License
 Nephra v1 falls under [META LLAMA 3 COMMUNITY LICENSE AGREEMENT](https://huggingface.co/meta-llama/Meta-Llama-3-8B/blob/main/LICENSE).