| { | |
| "best_metric": 33.39944765252322, | |
| "best_model_checkpoint": "/kaggle/working/whisper-small/checkpoint-1000", | |
| "epoch": 1.3245033112582782, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.033112582781456956, | |
| "grad_norm": 40.76376724243164, | |
| "learning_rate": 4.0000000000000003e-07, | |
| "loss": 4.304, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06622516556291391, | |
| "grad_norm": 34.9359245300293, | |
| "learning_rate": 9.000000000000001e-07, | |
| "loss": 3.9709, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09933774834437085, | |
| "grad_norm": 18.766077041625977, | |
| "learning_rate": 1.4000000000000001e-06, | |
| "loss": 3.4858, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.13245033112582782, | |
| "grad_norm": 16.810834884643555, | |
| "learning_rate": 1.9000000000000002e-06, | |
| "loss": 3.2396, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16556291390728478, | |
| "grad_norm": 19.717323303222656, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 3.1781, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.1986754966887417, | |
| "grad_norm": 17.67849349975586, | |
| "learning_rate": 2.9e-06, | |
| "loss": 3.2006, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.23178807947019867, | |
| "grad_norm": 16.326919555664062, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "loss": 3.146, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.26490066225165565, | |
| "grad_norm": 17.752124786376953, | |
| "learning_rate": 3.900000000000001e-06, | |
| "loss": 3.1915, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2980132450331126, | |
| "grad_norm": 17.477537155151367, | |
| "learning_rate": 4.4e-06, | |
| "loss": 3.0694, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.33112582781456956, | |
| "grad_norm": 17.810102462768555, | |
| "learning_rate": 4.9000000000000005e-06, | |
| "loss": 3.0796, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.36423841059602646, | |
| "grad_norm": 18.251075744628906, | |
| "learning_rate": 5.400000000000001e-06, | |
| "loss": 3.0895, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.3973509933774834, | |
| "grad_norm": 17.90201187133789, | |
| "learning_rate": 5.9e-06, | |
| "loss": 2.9872, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4304635761589404, | |
| "grad_norm": 18.006668090820312, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 3.0425, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.46357615894039733, | |
| "grad_norm": 16.625247955322266, | |
| "learning_rate": 6.9e-06, | |
| "loss": 2.9893, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.4966887417218543, | |
| "grad_norm": 19.76040267944336, | |
| "learning_rate": 7.4e-06, | |
| "loss": 2.9231, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.5298013245033113, | |
| "grad_norm": 21.64992904663086, | |
| "learning_rate": 7.9e-06, | |
| "loss": 3.0319, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5629139072847682, | |
| "grad_norm": 18.87610626220703, | |
| "learning_rate": 8.400000000000001e-06, | |
| "loss": 2.9039, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.5960264900662252, | |
| "grad_norm": 18.467191696166992, | |
| "learning_rate": 8.900000000000001e-06, | |
| "loss": 2.9607, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.6291390728476821, | |
| "grad_norm": 17.798429489135742, | |
| "learning_rate": 9.4e-06, | |
| "loss": 2.9454, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.6622516556291391, | |
| "grad_norm": 18.094806671142578, | |
| "learning_rate": 9.9e-06, | |
| "loss": 2.9388, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6622516556291391, | |
| "eval_cer": 102.70901330655285, | |
| "eval_loss": 2.9817957878112793, | |
| "eval_runtime": 2198.2774, | |
| "eval_samples_per_second": 2.264, | |
| "eval_steps_per_second": 0.283, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.695364238410596, | |
| "grad_norm": 16.370372772216797, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 2.9295, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.7284768211920529, | |
| "grad_norm": 18.59728240966797, | |
| "learning_rate": 9.100000000000001e-06, | |
| "loss": 2.9214, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.7615894039735099, | |
| "grad_norm": 16.144062042236328, | |
| "learning_rate": 8.6e-06, | |
| "loss": 2.9778, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.7947019867549668, | |
| "grad_norm": 17.924297332763672, | |
| "learning_rate": 8.1e-06, | |
| "loss": 2.8858, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8278145695364238, | |
| "grad_norm": 17.39291000366211, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 2.8875, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.8609271523178808, | |
| "grad_norm": 17.19579315185547, | |
| "learning_rate": 7.100000000000001e-06, | |
| "loss": 2.9667, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.8940397350993378, | |
| "grad_norm": 17.705387115478516, | |
| "learning_rate": 6.600000000000001e-06, | |
| "loss": 2.9193, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.9271523178807947, | |
| "grad_norm": 18.197153091430664, | |
| "learning_rate": 6.1e-06, | |
| "loss": 2.8616, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.9602649006622517, | |
| "grad_norm": 16.632579803466797, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 2.7889, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.9933774834437086, | |
| "grad_norm": 23.52530860900879, | |
| "learning_rate": 5.1e-06, | |
| "loss": 2.4231, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.0264900662251655, | |
| "grad_norm": 18.322874069213867, | |
| "learning_rate": 4.600000000000001e-06, | |
| "loss": 1.1389, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.0596026490066226, | |
| "grad_norm": 17.467435836791992, | |
| "learning_rate": 4.1e-06, | |
| "loss": 0.8244, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.0927152317880795, | |
| "grad_norm": 23.982784271240234, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "loss": 0.6974, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.1258278145695364, | |
| "grad_norm": 15.57445240020752, | |
| "learning_rate": 3.1000000000000004e-06, | |
| "loss": 0.6625, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.1589403973509933, | |
| "grad_norm": 16.174701690673828, | |
| "learning_rate": 2.6e-06, | |
| "loss": 0.6145, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.1920529801324504, | |
| "grad_norm": 17.91811752319336, | |
| "learning_rate": 2.1000000000000002e-06, | |
| "loss": 0.5941, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.2251655629139073, | |
| "grad_norm": 16.29407501220703, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 0.6053, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.2582781456953642, | |
| "grad_norm": 13.573580741882324, | |
| "learning_rate": 1.1e-06, | |
| "loss": 0.5533, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.2913907284768211, | |
| "grad_norm": 14.268310546875, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 0.5446, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.3245033112582782, | |
| "grad_norm": 13.822824478149414, | |
| "learning_rate": 1.0000000000000001e-07, | |
| "loss": 0.5492, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.3245033112582782, | |
| "eval_cer": 33.39944765252322, | |
| "eval_loss": 0.6020073294639587, | |
| "eval_runtime": 2262.2462, | |
| "eval_samples_per_second": 2.2, | |
| "eval_steps_per_second": 0.275, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.3245033112582782, | |
| "step": 1000, | |
| "total_flos": 4.65727933108224e+18, | |
| "train_loss": 2.47559251499176, | |
| "train_runtime": 12073.0146, | |
| "train_samples_per_second": 1.325, | |
| "train_steps_per_second": 0.083 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1000, | |
| "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": 4.65727933108224e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |