| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 40, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.001454428258423816, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0002, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.0001440563508689233, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 1.894846757755783e-05, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 2.1656050853304146e-05, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.010110922731659625, | |
| "learning_rate": 9.980973490458728e-05, | |
| "loss": 0.0003, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.022255514895711447, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.0104, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.005307385654747276, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 0.0011, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.005664941570268046, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0008, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.0001810605938768314, | |
| "learning_rate": 9.53153893518325e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.4331626638994084e-05, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.00032943658064544964, | |
| "learning_rate": 9.09576022144496e-05, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.014369175888068157, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0017, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.007524650230255758, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0007, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 5.194252325111768e-05, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 1.03939419962215e-05, | |
| "learning_rate": 7.86788218175523e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 4.97855079211176e-05, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.004178574654084881, | |
| "learning_rate": 7.113091308703498e-05, | |
| "loss": 0.0001, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 1.5480084175105143e-05, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 5.4352022955918e-05, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.0034842902670491043, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.0001, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.9281, | |
| "eval_samples_per_second": 1.078, | |
| "eval_steps_per_second": 1.078, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 0.0003737613969605659, | |
| "learning_rate": 5.435778713738292e-05, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.0006870061703080538, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 3.4683990021195124e-05, | |
| "learning_rate": 4.564221286261709e-05, | |
| "loss": 0.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 1.5959310174917984e-05, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 7.070564786027958e-05, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 3.050520547904037e-07, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 0.02863257382851171, | |
| "learning_rate": 2.886908691296504e-05, | |
| "loss": 0.0014, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 2.4742393963921362e-05, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 4.101310192512185e-05, | |
| "learning_rate": 2.132117818244771e-05, | |
| "loss": 0.0, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 2.581226977814674e-06, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 0.00028490599840642525, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.00018232442946056936, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 0.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 0.00012800637486506386, | |
| "learning_rate": 9.042397785550405e-06, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 0.00043110254240359794, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 0.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.0005447263675738613, | |
| "learning_rate": 4.684610648167503e-06, | |
| "loss": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 5.515463823856325e-06, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 2.2283685064171857e-05, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 0.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 1.0431202098647983e-05, | |
| "learning_rate": 7.596123493895991e-07, | |
| "loss": 0.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 6.800384630176213e-05, | |
| "learning_rate": 1.9026509541272275e-07, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.014197577325695893, | |
| "learning_rate": 0.0, | |
| "loss": 0.0008, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.48, | |
| "eval_samples_per_second": 0.676, | |
| "eval_steps_per_second": 0.676, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 40, | |
| "total_flos": 3851176427520.0, | |
| "train_loss": 0.0004419676075281487, | |
| "train_runtime": 303.2804, | |
| "train_samples_per_second": 0.264, | |
| "train_steps_per_second": 0.132 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 40, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3851176427520.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |