Safetensors
webshop / training

Commit History

  1. webshop_single_expert_qwen2_5_3_step_300_seed0: add best_val official eval (0.590/0.732)

    Verified signature
    07dbb25
  2. webshop_single_expert_qwen2_5_3_step_300_seed0: add best_val official eval (0.590/0.732)

    Verified signature
    9249583
  3. 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

    Verified signature
    cbff989

  1. Publish canonical sampler-fixed WebShop training checkpoints⁠

    Verified signature
    071f8fc