Text Generation
Transformers
Safetensors
PyTorch
English
gelu_gpt
gpt
gelu
261M
chinchilla
ablation
seed2
custom_code
Instructions to use mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch
- SGLang
How to use mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch 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 "mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch" \ --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": "mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch", "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 "mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch" \ --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": "mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch with Docker Model Runner:
docker model run hf.co/mlnomad/gelu-d12-chinchilla-261M-seed2-pytorch
File size: 1,394 Bytes
1a6203b | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 | """HuggingFace-compatible config for the flaxchat GELU GPT."""
from transformers import PretrainedConfig
class GeluGPTConfig(PretrainedConfig):
model_type = "gelu_gpt"
def __init__(
self,
sequence_len: int = 1024,
vocab_size: int = 32768,
n_layer: int = 12,
n_head: int = 12,
n_kv_head: int = 12,
n_embd: int = 768,
window_pattern: str = "SSSL",
tie_embeddings: bool = True,
rope_base: float = 100000.0,
pad_vocab_size_to: int = 64,
mlp: str = "gelu",
tie_word_embeddings: bool = True,
**kwargs,
):
self.sequence_len = sequence_len
self.vocab_size = vocab_size
self.n_layer = n_layer
self.n_head = n_head
self.n_kv_head = n_kv_head
self.n_embd = n_embd
self.window_pattern = window_pattern
self.tie_embeddings = tie_embeddings
self.rope_base = rope_base
self.pad_vocab_size_to = pad_vocab_size_to
self.mlp = mlp
self.max_position_embeddings = sequence_len * 10
# HF generation utilities probe these standard attribute names.
self.num_hidden_layers = n_layer
self.num_attention_heads = n_head
self.num_key_value_heads = n_kv_head
self.hidden_size = n_embd
super().__init__(tie_word_embeddings=tie_word_embeddings, **kwargs)
|