{ "model": "Qwen/Qwen2.5-3B", "renderer": "role_colon", "arms": { "rank1": { "get_info_at_creation": { "type": "get_info", "model_data": { "arch": null, "model_name": "Qwen/Qwen2.5-3B", "tokenizer_id": null }, "model_id": "276897f1a3424e2d2d784c743409490e", "optimizer_config": { "type": "adamw" }, "is_lora": true, "lora_rank": 1, "model_name": "Qwen/Qwen2.5-3B" }, "initial_output_sha256": "c3ff0be48f064881c30bbeb9c7454d22ad3f4749393ff56c94ff36c6bfb96731", "n_logprobs": 421, "adapter_error": "attempt 1: save_weights_external -> 404 Not Found on shared Spindle (ev/probe_attempt1.log); not retried", "fwd_bwd_metrics": { "loss:mean": -0.2890995260663507, "tokens:sum": 211.0, "loss:sum": -61.0, "response_length:mean": 211.0, "n_sequences:sum": 1.0 }, "optim_step_metrics": { "update_successful:mean": 1.0, "timing/optimizer_step": 0.0, "timing/idle_wait_s": 5.4397311029999855, "timing/idle_wait_calls": 7.0, "timing/prepare_batch_s": 0.00038183300000582676, "timing/prepare_batch_calls": 2.0, "timing/forward_only_s": 0.2164900980000084, "timing/forward_only_calls": 1.0, "timing/build_outputs_s": 0.0002101219999985915, "timing/build_outputs_calls": 2.0, "timing/forward_backward_s": 1.5641870960000688, "timing/forward_backward_calls": 1.0, "timing/optim_step_s": 0.024110317000008763, "timing/optim_step_calls": 1.0, "timing/trainer_step_s": 1.5882974130000775, "grad_norm:mean": 744.4153442382812 }, "get_info_after_step": { "type": "get_info", "model_data": { "arch": null, "model_name": "Qwen/Qwen2.5-3B", "tokenizer_id": null }, "model_id": "276897f1a3424e2d2d784c743409490e", "optimizer_config": { "type": "adamw" }, "is_lora": true, "lora_rank": 1, "model_name": "Qwen/Qwen2.5-3B" }, "post_step_output_sha256": "647060c02b081460d602fa057260bc9590e818e68620cf97e9c3d136154590dd", "post_step_max_abs_logprob_change": 6.993038456887007 }, "rank32": { "get_info_at_creation": { "type": "get_info", "model_data": { "arch": null, "model_name": "Qwen/Qwen2.5-3B", "tokenizer_id": null }, "model_id": "9f734c17eaaf2ef515cec680f0a3782a", "optimizer_config": { "type": "adamw" }, "is_lora": true, "lora_rank": 32, "model_name": "Qwen/Qwen2.5-3B" }, "initial_output_sha256": "c3ff0be48f064881c30bbeb9c7454d22ad3f4749393ff56c94ff36c6bfb96731", "n_logprobs": 421, "adapter_error": "attempt 1: save_weights_external -> 404 Not Found on shared Spindle (ev/probe_attempt1.log); not retried", "fwd_bwd_metrics": { "loss:mean": -0.2890995260663507, "tokens:sum": 211.0, "loss:sum": -61.0, "response_length:mean": 211.0, "n_sequences:sum": 1.0 }, "optim_step_metrics": { "update_successful:mean": 1.0, "timing/optimizer_step": 0.0, "timing/idle_wait_s": 3.0118267169999626, "timing/idle_wait_calls": 7.0, "timing/prepare_batch_s": 0.0004170030000523184, "timing/prepare_batch_calls": 2.0, "timing/forward_only_s": 0.151421809999988, "timing/forward_only_calls": 1.0, "timing/build_outputs_s": 0.00018185400006132113, "timing/build_outputs_calls": 2.0, "timing/forward_backward_s": 0.38535455600003843, "timing/forward_backward_calls": 1.0, "timing/optim_step_s": 0.036620599000002585, "timing/optim_step_calls": 1.0, "timing/trainer_step_s": 0.421975155000041, "grad_norm:mean": 120.1463623046875 }, "get_info_after_step": { "type": "get_info", "model_data": { "arch": null, "model_name": "Qwen/Qwen2.5-3B", "tokenizer_id": null }, "model_id": "9f734c17eaaf2ef515cec680f0a3782a", "optimizer_config": { "type": "adamw" }, "is_lora": true, "lora_rank": 32, "model_name": "Qwen/Qwen2.5-3B" }, "post_step_output_sha256": "b87e69f79cc778e929a485ce50876e8386d478964ae3bc9cfb9927a16a55e7d2", "post_step_max_abs_logprob_change": 7.007339611649513 } }, "fixed_batch_sha256": "bd6e130ab84c8977aea60bc4571297ae3a3a090eb9d9463fcbc3439214bfcb3e", "initial_outputs_match": true, "initial_logprob_max_abs_diff": 0.0, "seconds": 31.080610275268555 }