Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets new
  • Docs
  • Enterprise
  • Pricing

  • Log In
  • Sign Up

s0wa48
/
sn97-commit-mantis7

Safetensors
qwen3_5_text
Model card Files Files and versions
xet
Community
sn97-commit-mantis7
8.43 GB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 3 commits
s0wa48's picture
s0wa48
merge α=0.7 (70% tied + 30% GKD): kl=0.120 (−26% vs king), cap=0.861, rkl=0.332
9f03d2b verified 20 days ago
  • .gitattributes
    1.57 kB
    initial commit: tied_baseline sparse128-30k/best, Arbos fair eval avg Δ=-41.2% vs king on 3 blocks 20 days ago
  • chat_template.jinja
    7.76 kB
    initial commit: tied_baseline sparse128-30k/best, Arbos fair eval avg Δ=-41.2% vs king on 3 blocks 20 days ago
  • config.json
    2.36 kB
    initial commit: tied_baseline sparse128-30k/best, Arbos fair eval avg Δ=-41.2% vs king on 3 blocks 20 days ago
  • config.json.bak
    1.98 kB
    initial commit: tied_baseline sparse128-30k/best, Arbos fair eval avg Δ=-41.2% vs king on 3 blocks 20 days ago
  • generation_config.json
    120 Bytes
    initial commit: tied_baseline sparse128-30k/best, Arbos fair eval avg Δ=-41.2% vs king on 3 blocks 20 days ago
  • model.safetensors
    8.41 GB
    xet
    merge α=0.7 (70% tied + 30% GKD): kl=0.120 (−26% vs king), cap=0.861, rkl=0.332 20 days ago
  • preprocessor_config.json
    168 Bytes
    initial commit: tied_baseline sparse128-30k/best, Arbos fair eval avg Δ=-41.2% vs king on 3 blocks 20 days ago
  • tokenizer.json
    20 MB
    xet
    initial commit: tied_baseline sparse128-30k/best, Arbos fair eval avg Δ=-41.2% vs king on 3 blocks 20 days ago
  • tokenizer_config.json
    1.1 kB
    initial commit: tied_baseline sparse128-30k/best, Arbos fair eval avg Δ=-41.2% vs king on 3 blocks 20 days ago