logan7000/mllm-cogrpo-heter-qwen25vl-7b-x-internvl35-8b-mmr1-old3ep-b16-groupB-endpoint

optimizer states (global_step1083) for optimizer

Upload trainer_state.json with huggingface_hub

Upload trainer_state_full.json with huggingface_hub

co-old3ep batch-16 InternVL endpoint s1083 (internal judged 70.93) + training log

initial commit