Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
gptoss120b-sdpo-exp10c-argmax-revkl-floor10-step320 – AI Model by JacopoCirica | AlphaNeural AI
You can deploy this model and start earning money today!
JacopoCirica
/
gptoss120b-sdpo-exp10c-argmax-revkl-floor10-step320
like
0
safetensors
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
gpt-oss-120b SDPO++ Exp 10c step-320 (epoch3, adv-recovered) raw Tinker LoRA adapter
argmax+revKL, teacher_lp_floor=-10, MCP-480. r=32 alpha=32 all-linear. Merge into base at deploy (build_hf_model, merge_strategy=auto, MXFP4-preserving).
