Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets
  • Docs
  • Enterprise
  • Pricing
    • Website
      • Tasks
      • HuggingChat
      • Collections
      • Languages
      • Organizations
    • Community
      • Blog
      • Posts
      • Daily Papers
      • Hardware
      • Learn
      • Discord
      • Forum
      • GitHub
    • Solutions
      • Team & Enterprise
      • Hugging Face PRO
      • Enterprise Support
      • Inference Providers
      • Inference Endpoints
      • Storage Buckets

  • Log In
  • Sign Up

Spaces:
philippyt
/
ablation-lab
Running

App Files Files Community
Fetching metadata from the HF Docker repository...
ablation-lab / training
129 kB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 2 commits
philippyt's picture
philippyt
setting up training kind A and kind B sets, improvements to frontend
e7b5745 6 months ago
  • configs
    setting up training kind A and kind B sets, improvements to frontend 6 months ago
  • __init__.py
    0 Bytes
    MVP1: working prototype of transformer ablation playground. web app for composing tiny transformer architectures, running real inference, and seeing how each architectural switch affects cost and quality. Hash-stable arch_config shared between Python and TS; closed-form cost formulas parity-tested against actual sum(p.numel()) 6 months ago
  • data.py
    971 Bytes
    MVP1: working prototype of transformer ablation playground. web app for composing tiny transformer architectures, running real inference, and seeing how each architectural switch affects cost and quality. Hash-stable arch_config shared between Python and TS; closed-form cost formulas parity-tested against actual sum(p.numel()) 6 months ago
  • seed.py
    282 Bytes
    MVP1: working prototype of transformer ablation playground. web app for composing tiny transformer architectures, running real inference, and seeing how each architectural switch affects cost and quality. Hash-stable arch_config shared between Python and TS; closed-form cost formulas parity-tested against actual sum(p.numel()) 6 months ago
  • train.py
    5.96 kB
    setting up training kind A and kind B sets, improvements to frontend 6 months ago