Hugging Face's logo

JohnP1
/
d1a-e4b

Text Classification
PEFT
Safetensors
English
Japanese
d1a
decision-model
calibration
lora
gemma4
typesafe
system-one
pull-requests
Model card Files
xet
 Community
d1a-e4b / training_config.json

Commit History

10 days ago

  1. D1A-E4B v0.6

    JohnP1
    Verified signature
    3ca01b8

13 days ago

  1. D1A-E4B v0.5: Kev's skills back on top of v0.4 (hard 55->71, devtools 64->70); PR blast/severity trade-off in the card

    JohnP1
    Verified signature
    e7db32f

17 days ago

  1. v0.4: pull-request labeling round 2

    JohnP1
    Verified signature
    dbd2df8

18 days ago

  1. v0.3: pull-request labeling (English and Japanese)

    JohnP1
    Verified signature
    c9bf9f6

  2. D1A-E4B v0.2 (hybrid): + Kev recipe stages + Japanese JGLUE + agent-factory/routing; calibrated T=1.95

    JohnP1
    Verified signature
    e8064af

19 days ago

  1. D1A-E4B v0.1: two epochs at lr 5e-5 (dev acc 0.853 trained / 0.680 new sources)

    JohnP1
    Verified signature
    98847a9