Views
No views yet
| Component | Value |
|---|---|
| Hidden Size | 768 |
| Layers | 12 |
| Attention Heads | 12 |
| Field Regions | 128 |
| Field Size | 128 |
| Field Depth | 64 |
┌─────────────────────────────────────┐
│ TRANSFORMER ENCODER LAYERS │
│ (Self-Attention + FFN) │
└─────────────────────────────────────┘
↓
┌─────────────────────────────────────┐
│ FIELD DYNAMICS CORE │
│ (Sparse Activation + Evolution) │
└─────────────────────────────────────┘
↓
┌─────────────────────────────────────┐
│ OUTPUT PROJECTION │
│ (Pattern → Token Logits) │
└─────────────────────────────────────┘| File | Description |
|---|---|
Model-001.safetensors | Model weights (721.30 MB) |
config.json | Model configuration |
tokenizer.json | Tokenizer vocabulary |
tokenizer_config.json | Tokenizer configuration |
generation_config.json | Generation parameters |
params.json | LIGHTBRAIN parameters |
1from lightbrain.model import HybridFieldTransformer
2from lightbrain.inference import InferenceEngine
3
4# Load model
5model = HybridFieldTransformer.load("path/to/model")
6engine = InferenceEngine(model=model)
7
8# Generate
9result = engine.generate("Hello, how are you?")
10print(result.text)1from safetensors.numpy import load_file
2import json
3
4# Load weights
5weights = load_file("Model-001.safetensors")
6
7# Load config
8with open("config.json") as f:
9 config = json.load(f)
10
11# Reconstruct model from weights1# Install
2!pip install safetensors
3
4# Download
5from huggingface_hub import snapshot_download
6model_path = snapshot_download(repo_id="lightbrain-100m")
7
8# Load and use
9from safetensors.numpy import load_file
10weights = load_file(f"{model_path}/Model-001.safetensors")1@misc{lightbrain2024,
2 title={LIGHTBRAIN: Hybrid Field Dynamics for Efficient LLMs},
3 year={2024},
4 publisher={HuggingFace}
5}