webshop_single_expert_qwen2_5_3_step_300_seed0: add best_val official eval (0.590/0.732) 07dbb25 verified Davidk1 commited on Aug 27
webshop_single_expert_qwen2_5_3_step_300_seed0: add best_val official eval (0.590/0.732) 9249583 verified Davidk1 commited on Aug 27
webshop_single_expert_qwen2_5_3_step_300_seed0: LoRA adapters (steps 100/200/300, best_val) + manifest/README; first 3B run, below 1.5B reference cbff989 verified Davidk1 commited on Aug 27
Publish canonical sampler-fixed WebShop training checkpoints 071f8fc verified PengxinWang commited on Aug 26
Add sampler-fixed WebShop step-300 single- and 2-expert checkpoints e7966a1 verified PengxinWang commited on Aug 24
Add 2expert clusterF nothinking best (step500, val SR 0.742) be2c663 verified Davidk1 commited on Aug 19
Add 2expert mlp random: best==last==step1000 (val SR 0.758) d29a42a verified Davidk1 commited on Aug 19