| { | |
| "best_metric": 21.28044187798142, | |
| "best_model_checkpoint": "/kaggle/working/whisper-small/checkpoint-2000", | |
| "epoch": 2.6490066225165565, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "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": 1.3576158940397351, | |
| "grad_norm": 15.552218437194824, | |
| "learning_rate": 6.540000000000001e-06, | |
| "loss": 0.501, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.390728476821192, | |
| "grad_norm": 18.500070571899414, | |
| "learning_rate": 6.373333333333334e-06, | |
| "loss": 0.5625, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.423841059602649, | |
| "grad_norm": 13.384895324707031, | |
| "learning_rate": 6.206666666666668e-06, | |
| "loss": 0.4785, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 1.4569536423841059, | |
| "grad_norm": 11.979981422424316, | |
| "learning_rate": 6.040000000000001e-06, | |
| "loss": 0.4728, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.490066225165563, | |
| "grad_norm": 13.125775337219238, | |
| "learning_rate": 5.873333333333334e-06, | |
| "loss": 0.4484, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 1.5231788079470199, | |
| "grad_norm": 10.14282512664795, | |
| "learning_rate": 5.713333333333334e-06, | |
| "loss": 0.4227, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.5562913907284768, | |
| "grad_norm": 14.51291561126709, | |
| "learning_rate": 5.546666666666667e-06, | |
| "loss": 0.4419, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 1.589403973509934, | |
| "grad_norm": 13.863635063171387, | |
| "learning_rate": 5.380000000000001e-06, | |
| "loss": 0.4429, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.6225165562913908, | |
| "grad_norm": 11.522802352905273, | |
| "learning_rate": 5.213333333333334e-06, | |
| "loss": 0.3981, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 1.6556291390728477, | |
| "grad_norm": 13.733115196228027, | |
| "learning_rate": 5.046666666666668e-06, | |
| "loss": 0.3974, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.6887417218543046, | |
| "grad_norm": 11.775789260864258, | |
| "learning_rate": 4.880000000000001e-06, | |
| "loss": 0.3978, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 1.7218543046357615, | |
| "grad_norm": 8.765869140625, | |
| "learning_rate": 4.713333333333334e-06, | |
| "loss": 0.3812, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.7549668874172184, | |
| "grad_norm": 10.572060585021973, | |
| "learning_rate": 4.546666666666667e-06, | |
| "loss": 0.4257, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 1.7880794701986755, | |
| "grad_norm": 11.202813148498535, | |
| "learning_rate": 4.38e-06, | |
| "loss": 0.3635, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.8211920529801324, | |
| "grad_norm": 10.402491569519043, | |
| "learning_rate": 4.213333333333333e-06, | |
| "loss": 0.3636, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 1.8543046357615895, | |
| "grad_norm": 22.270811080932617, | |
| "learning_rate": 4.046666666666667e-06, | |
| "loss": 0.3696, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.8874172185430464, | |
| "grad_norm": 8.238192558288574, | |
| "learning_rate": 3.88e-06, | |
| "loss": 0.397, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 1.9205298013245033, | |
| "grad_norm": 9.551776885986328, | |
| "learning_rate": 3.713333333333334e-06, | |
| "loss": 0.3747, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.9536423841059603, | |
| "grad_norm": 13.149867057800293, | |
| "learning_rate": 3.5466666666666673e-06, | |
| "loss": 0.3771, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 1.9867549668874172, | |
| "grad_norm": 13.081770896911621, | |
| "learning_rate": 3.3800000000000007e-06, | |
| "loss": 0.352, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.9867549668874172, | |
| "eval_cer": 23.299020838563898, | |
| "eval_loss": 0.4214184284210205, | |
| "eval_runtime": 2258.7536, | |
| "eval_samples_per_second": 2.203, | |
| "eval_steps_per_second": 0.275, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.019867549668874, | |
| "grad_norm": 7.450255393981934, | |
| "learning_rate": 3.213333333333334e-06, | |
| "loss": 0.2911, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 2.052980132450331, | |
| "grad_norm": 8.604903221130371, | |
| "learning_rate": 3.0466666666666666e-06, | |
| "loss": 0.2069, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.0860927152317883, | |
| "grad_norm": 5.367754936218262, | |
| "learning_rate": 2.88e-06, | |
| "loss": 0.2461, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 2.119205298013245, | |
| "grad_norm": 11.53250789642334, | |
| "learning_rate": 2.7133333333333333e-06, | |
| "loss": 0.2071, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.152317880794702, | |
| "grad_norm": 9.057580947875977, | |
| "learning_rate": 2.5466666666666667e-06, | |
| "loss": 0.2255, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 2.185430463576159, | |
| "grad_norm": 8.253719329833984, | |
| "learning_rate": 2.38e-06, | |
| "loss": 0.1968, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.218543046357616, | |
| "grad_norm": 10.867476463317871, | |
| "learning_rate": 2.2133333333333335e-06, | |
| "loss": 0.2177, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 2.251655629139073, | |
| "grad_norm": 10.779939651489258, | |
| "learning_rate": 2.046666666666667e-06, | |
| "loss": 0.2191, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.2847682119205297, | |
| "grad_norm": 8.389144897460938, | |
| "learning_rate": 1.8800000000000002e-06, | |
| "loss": 0.2137, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 2.3178807947019866, | |
| "grad_norm": 11.38824462890625, | |
| "learning_rate": 1.7133333333333336e-06, | |
| "loss": 0.2326, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.3509933774834435, | |
| "grad_norm": 10.286229133605957, | |
| "learning_rate": 1.546666666666667e-06, | |
| "loss": 0.2014, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 2.384105960264901, | |
| "grad_norm": 9.5010986328125, | |
| "learning_rate": 1.3800000000000001e-06, | |
| "loss": 0.2121, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.4172185430463577, | |
| "grad_norm": 10.394664764404297, | |
| "learning_rate": 1.2133333333333335e-06, | |
| "loss": 0.2166, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 2.4503311258278146, | |
| "grad_norm": 13.732166290283203, | |
| "learning_rate": 1.0466666666666669e-06, | |
| "loss": 0.2243, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 2.4834437086092715, | |
| "grad_norm": 7.431657791137695, | |
| "learning_rate": 8.8e-07, | |
| "loss": 0.2186, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 2.5165562913907285, | |
| "grad_norm": 5.561306476593018, | |
| "learning_rate": 7.133333333333334e-07, | |
| "loss": 0.1992, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.5496688741721854, | |
| "grad_norm": 8.894646644592285, | |
| "learning_rate": 5.466666666666667e-07, | |
| "loss": 0.2195, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 2.5827814569536423, | |
| "grad_norm": 8.13320541381836, | |
| "learning_rate": 3.8e-07, | |
| "loss": 0.2235, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.6158940397350996, | |
| "grad_norm": 8.711597442626953, | |
| "learning_rate": 2.1333333333333334e-07, | |
| "loss": 0.2033, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 2.6490066225165565, | |
| "grad_norm": 9.557293891906738, | |
| "learning_rate": 4.6666666666666674e-08, | |
| "loss": 0.2243, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.6490066225165565, | |
| "eval_cer": 21.28044187798142, | |
| "eval_loss": 0.38667094707489014, | |
| "eval_runtime": 2236.5916, | |
| "eval_samples_per_second": 2.225, | |
| "eval_steps_per_second": 0.278, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.6490066225165565, | |
| "step": 2000, | |
| "total_flos": 9.31455866216448e+18, | |
| "train_loss": 0.15934584045410155, | |
| "train_runtime": 12035.0029, | |
| "train_samples_per_second": 2.659, | |
| "train_steps_per_second": 0.166 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 9.31455866216448e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |