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

  • Log In
  • Sign Up

cuber12
/
dropout-decay

English
dropout
streaming
language-modeling
transformer
mps
reproducibility
Model card Files Files and versions
xet
Community
dropout-decay / scripts
58 kB
Ctrl+K
Ctrl+K
  • 2 contributors
History: 7 commits
Mandeep Sidhu
Use absolute regime names for streaming reports
dcae82e 3 days ago
  • fit_dropout_coefficients.py
    21.6 kB
    Add dropout pressure validation artifacts 3 days ago
  • make_streaming_anchors.py
    4 kB
    Add dropout pressure validation artifacts 3 days ago
  • prepare_tinystories.py
    2.12 kB
    Add TinyStories corpus holdout results 5 days ago
  • prepare_wikitext103.py
    2.17 kB
    Add reproducible WikiText corpus prep 5 days ago
  • run_experiments.py
    774 Bytes
    Initial dropout decay research pipeline 8 days ago
  • summarize_cross_regime_backtest.py
    12.2 kB
    Use absolute regime names for streaming reports 3 days ago
  • summarize_streaming_multiseed.py
    15 kB
    Add previous local streaming report 3 days ago