Instructions to use thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21 with Transformers:

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

pipe = pipeline("text-generation", model="thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21

SGLang

How to use thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21 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 "thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21" \
    --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": "thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21",
		"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 "thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21" \
        --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": "thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21 with Docker Model Runner:
```
docker model run hf.co/thrunlab/sparse_llama_debugging_refined_web_90p_debugging_2024-03-21
```

sparse_llama_debugging_refined_web_90p_debugging_2024-03-21 / config.json

lukeleeai

End of training

58cd674 verified over 2 years ago

Raw

History Blame Contribute Delete

1.06 kB

	{
	"architectures": [
	"SparseLlamaForCausalLM"
	],
	"attention_bias": false,
	"attention_dropout": 0.0,
	"auto_map": {
	"AutoConfig": "ugly_utils.SparseLlamaConfig",
	"AutoModelForCausalLM": "ugly_utils.SparseLlamaForCausalLM"
	},
	"bos_token_id": 1,
	"eos_token_id": 2,
	"hidden_act": "silu",
	"hidden_size": 64,
	"initializer_range": 0.02,
	"intermediate_size": 64,
	"max_position_embeddings": 2048,
	"model_type": "sparse_llama",
	"num_attention_heads": 32,
	"num_hidden_layers": 4,
	"num_key_value_heads": 32,
	"pretraining_tp": 1,
	"rms_norm_eps": 1e-06,
	"rope_scaling": null,
	"rope_theta": 10000.0,
	"thresholds": [
	0.12938815355300903,
	0.12938815355300903,
	0.1313941776752472,
	0.12337010353803635
	],
	"tie_word_embeddings": false,
	"torch_dtype": "float32",
	"transformers_version": "4.37.2",
	"use_cache": true,
	"use_graceful_regularization": false,
	"use_relu": false,
	"use_sparse_model": true,
	"use_sparse_predictor": false,
	"use_sparse_regularization": false,
	"vocab_size": 32000
	}