wire-9m-best / README.md
LisaMegaWatts's picture
Add Wire-9M (H32 L8) checkpoint (step 5000, bpb 3.0866)
4b8792b verified
metadata
license: mit
tags:
  - harmonic-gpt
  - oscillator-dynamics
  - clifford-algebra
  - byte-level
  - from-scratch
  - monumental-systems
language: en
pipeline_tag: text-generation

Wire-9M (H32 L8)

WireNative 9M — n_harmonics=32, 8 layers, best BPB checkpoint

Part of the Harmonic GPT research into oscillator-based neural computation.

Architecture: WireNative

Property Value
Parameters 8,894,520
BPB 3.0866
Training step 5,000
n_harmonics 64
n_layers 8
n_groups 7
d_model 896
Vocab 256 (raw bytes)

Usage

from huggingface_hub import hf_hub_download
from safetensors.torch import load_file

weights = load_file(hf_hub_download("MonumentalSystems/wire-9m-best", "model.safetensors"))
config = json.load(open(hf_hub_download("MonumentalSystems/wire-9m-best", "config.json")))

All operations are native Clifford algebra / harmonic oscillator dynamics — no softmax attention, no MLP, no ReLU.