Instructions to use Lanni-ni/alibi_babylm_100m_2layer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Lanni-ni/alibi_babylm_100m_2layer with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Lanni-ni/alibi_babylm_100m_2layer", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("Lanni-ni/alibi_babylm_100m_2layer", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Lanni-ni/alibi_babylm_100m_2layer with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Lanni-ni/alibi_babylm_100m_2layer" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Lanni-ni/alibi_babylm_100m_2layer", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Lanni-ni/alibi_babylm_100m_2layer
- SGLang
How to use Lanni-ni/alibi_babylm_100m_2layer 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 "Lanni-ni/alibi_babylm_100m_2layer" \ --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": "Lanni-ni/alibi_babylm_100m_2layer", "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 "Lanni-ni/alibi_babylm_100m_2layer" \ --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": "Lanni-ni/alibi_babylm_100m_2layer", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Lanni-ni/alibi_babylm_100m_2layer with Docker Model Runner:
docker model run hf.co/Lanni-ni/alibi_babylm_100m_2layer
File size: 1,257 Bytes
59e90da | 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 43 44 | import torch
import torch.nn
from typing import Dict, Any
class LayerWithVisualization(torch.nn.Module):
def __init__(self):
super().__init__()
self.visualization_enabled = False
def prepare(self):
# Should be called before the training step
pass
def plot(self, options: Dict[str, Any]) -> Dict[str, Any]:
raise NotImplementedError()
class LayerVisualizer:
def __init__(self, module: torch.nn.Module, options: Dict[str, Any] = {}):
self.modules = []
self.options = options
self.curr_options = None
for n, m in module.named_modules():
if isinstance(m, LayerWithVisualization):
self.modules.append((n, m))
def plot(self) -> Dict[str, Any]:
res = {}
for n, m in self.modules:
res.update({f"{n}/{k}": v for k, v in m.plot(self.curr_options).items()})
m.visualization_enabled = False
self.curr_options = None
return res
def prepare(self, options: Dict[str, Any] = {}):
self.curr_options = self.options.copy()
self.curr_options.update(options)
for _, m in self.modules:
m.prepare()
m.visualization_enabled = True
|