Instructions to use gradientai/Llama-3-70B-Instruct-Gradient-262k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use gradientai/Llama-3-70B-Instruct-Gradient-262k with Transformers:

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

pipe = pipeline("text-generation", model="gradientai/Llama-3-70B-Instruct-Gradient-262k")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("gradientai/Llama-3-70B-Instruct-Gradient-262k")
model = AutoModelForCausalLM.from_pretrained("gradientai/Llama-3-70B-Instruct-Gradient-262k", 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 gradientai/Llama-3-70B-Instruct-Gradient-262k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "gradientai/Llama-3-70B-Instruct-Gradient-262k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "gradientai/Llama-3-70B-Instruct-Gradient-262k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/gradientai/Llama-3-70B-Instruct-Gradient-262k

SGLang

How to use gradientai/Llama-3-70B-Instruct-Gradient-262k 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 "gradientai/Llama-3-70B-Instruct-Gradient-262k" \
    --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": "gradientai/Llama-3-70B-Instruct-Gradient-262k",
		"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 "gradientai/Llama-3-70B-Instruct-Gradient-262k" \
        --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": "gradientai/Llama-3-70B-Instruct-Gradient-262k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use gradientai/Llama-3-70B-Instruct-Gradient-262k with Docker Model Runner:
```
docker model run hf.co/gradientai/Llama-3-70B-Instruct-Gradient-262k
```

forrest-gradient commited on May 3, 2024

Commit

e77cde0

verified ·

1 Parent(s): 2f636d9

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -2

README.md CHANGED Viewed

@@ -42,13 +42,12 @@ For training data, we generate long contexts by augmenting [SlimPajama](https://
 | Initialize From          | Llama-3-70B-Instruct             | 65K            |
 | Sequence Length 2^N      | 16              | 18              |
 | RoPE theta               | 15,296,098      | 207,112,184     |
-| Batch Size               | 1               | 1               |
 | Gradient Accumulation Steps | 1           | 1               |
 | Steps                    | 20              | 25              |
 | Total Tokens             | 83,886,080      | 104,857,600     |
 | Learning rate            | 0.00002         | 0.00002         |
 | # GPUs                   | 512             | 512             |
-| Ring parallelism         | 64              | 16              |
 | GPU Type                 | NVIDIA L40S     | NVIDIA L40S     |
 | Minutes to Train (Wall)  | 100             | 170             |

 | Initialize From          | Llama-3-70B-Instruct             | 65K            |
 | Sequence Length 2^N      | 16              | 18              |
 | RoPE theta               | 15,296,098      | 207,112,184     |
+| Batch Size               | 64               | 16               |
 | Gradient Accumulation Steps | 1           | 1               |
 | Steps                    | 20              | 25              |
 | Total Tokens             | 83,886,080      | 104,857,600     |
 | Learning rate            | 0.00002         | 0.00002         |
 | # GPUs                   | 512             | 512             |
 | GPU Type                 | NVIDIA L40S     | NVIDIA L40S     |
 | Minutes to Train (Wall)  | 100             | 170             |