{ "init": { "base_model": "Qwen/Qwen3.5-2B", "rank": 32, "seed": null, "data_seed": 0, "init_state_path": "tinker://21e32976f18f971463410adff4e364e4/weights/rx19badd842c_init", "model_id": "21e32976f18f971463410adff4e364e4", "model_data": { "arch": null, "model_name": "Qwen/Qwen3.5-2B", "tokenizer_id": null }, "optimizer_config": { "type": "adamw" }, "seconds": 13.081110715866089 }, "pilot_restore": { "arm": "dppo_pilot", "loss": "dppo", "restored_from": "tinker://21e32976f18f971463410adff4e364e4/weights/rx19badd842c_init", "restore_count": 1, "model_id": "8004a92b-0b51-4ec0-a996-ab27ef0a7254:train:0", "executed_base_model": "Qwen/Qwen3.5-2B", "optimizer_after_restore": { "type": "adamw" }, "optimizer_step_params": { "learning_rate": 2e-05, "beta1": 0.9, "beta2": 0.95, "eps": 1e-08, "weight_decay": 0.0, "grad_clip_norm": 0.0 }, "fixed_batch_sha256": "5c545b6102e24b6f68b2a5e593ad2f7a81d85c4c170b97533ac4bc0f494d772b", "init_fixed_logprobs_sha256": "10572d61d17d3ec76f755aa2123b85033e381e883af6983427af9fb01edab837", "init_fixed_logprobs_n": 2206, "before_first_update": true, "setup_seconds": 46.56659531593323 }, "pilot_final": { "steps": 5, "sampler_path": "tinker://8004a92b-0b51-4ec0-a996-ab27ef0a7254:train:0/sampler_weights/dppo_pilot_final_s5", "train_seconds": 354.9852879047394 }, "pilot_last_state": { "step": 5, "state_path": "tinker://8004a92b-0b51-4ec0-a996-ab27ef0a7254:train:0/weights/dppo_pilot_state_s5" } }