gigpo-sft-w3c20-sync1-0726-step50 – AI Model by yinita | AlphaNeural AI

You can deploy this model and start earning money today!

yinita/
gigpo-sft-w3c20-sync1-0726-step50

like

0

gigpo-sft-w3c20-sync1-0726

yinita/ps4mas-sft-x5-single-ep3

Views

No views yet

gigpo-sft-w3c20-sync1-0726 — step 50

LoRA adapter from GiGPO run gigpo_sft_w3c20_sync1_50_0726_0459 at step 50.

Run config

Base: yinita/ps4mas-sft-x5-single-ep3
Judge: Claude Sonnet 4.6 (Bedrock)
Reward: phase2-only (acq_w=3, comp_w=20), reward_phase1_steps=0
sync-every=1, force_acquire_frac=0, max_steps=50
WandB: https://wandb.ai/yinita/PS4MAS_GiGPO/runs/cgtq25cv

Case study @ step 50 (n=10, Sonnet)

mean_composite: 2.499
acquire_rate: 0.10

Series

repo	step	CS composite	CS acq
`yinita/gigpo-sft-w3c20-sync1-0726-step10`	10	3.066	0.50
`yinita/gigpo-sft-w3c20-sync1-0726-step20`	20	2.732	0.40
`yinita/gigpo-sft-w3c20-sync1-0726-step30`	30	2.532	0.20
`yinita/gigpo-sft-w3c20-sync1-0726-step40`	40	2.600	0.10
`yinita/gigpo-sft-w3c20-sync1-0726-step50`	50	2.499	0.10

Best CS in this series: step10. Adapter-only (no optimizer). Load as LoRA on the SFT base (e.g. vLLM --enable-lora).