| { |
| "best_global_step": 2000, |
| "best_metric": 100.0, |
| "best_model_checkpoint": "./whisper-small-az/checkpoint-2000", |
| "epoch": 1.0, |
| "eval_steps": 2000, |
| "global_step": 157959, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 6.330756715350186e-06, |
| "grad_norm": 18.461782455444336, |
| "learning_rate": 0.0, |
| "loss": 0.6858, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.012661513430700372, |
| "grad_norm": 20.084333419799805, |
| "learning_rate": 4.952400307381604e-06, |
| "loss": 0.683, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.012661513430700372, |
| "eval_loss": 0.5455414652824402, |
| "eval_runtime": 164.1881, |
| "eval_samples_per_second": 1.218, |
| "eval_steps_per_second": 1.218, |
| "eval_wer": 100.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.025323026861400744, |
| "grad_norm": 15.010261535644531, |
| "learning_rate": 4.888891711493151e-06, |
| "loss": 0.4948, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.025323026861400744, |
| "eval_loss": 0.48741215467453003, |
| "eval_runtime": 97.6363, |
| "eval_samples_per_second": 2.048, |
| "eval_steps_per_second": 2.048, |
| "eval_wer": 100.0, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.03798454029210112, |
| "grad_norm": 17.379833221435547, |
| "learning_rate": 4.825383115604698e-06, |
| "loss": 0.4432, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.03798454029210112, |
| "eval_loss": 0.4483162760734558, |
| "eval_runtime": 181.4745, |
| "eval_samples_per_second": 1.102, |
| "eval_steps_per_second": 1.102, |
| "eval_wer": 100.0, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.05064605372280149, |
| "grad_norm": 26.042007446289062, |
| "learning_rate": 4.761874519716244e-06, |
| "loss": 0.4174, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.05064605372280149, |
| "eval_loss": 0.41587966680526733, |
| "eval_runtime": 111.3284, |
| "eval_samples_per_second": 1.796, |
| "eval_steps_per_second": 1.796, |
| "eval_wer": 100.0, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.06330756715350186, |
| "grad_norm": 23.43629264831543, |
| "learning_rate": 4.698365923827791e-06, |
| "loss": 0.3872, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.06330756715350186, |
| "eval_loss": 0.4035486578941345, |
| "eval_runtime": 97.0749, |
| "eval_samples_per_second": 2.06, |
| "eval_steps_per_second": 2.06, |
| "eval_wer": 100.0, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.07596908058420224, |
| "grad_norm": 15.88780403137207, |
| "learning_rate": 4.634857327939337e-06, |
| "loss": 0.3519, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.07596908058420224, |
| "eval_loss": 0.3873368203639984, |
| "eval_runtime": 127.0378, |
| "eval_samples_per_second": 1.574, |
| "eval_steps_per_second": 1.574, |
| "eval_wer": 100.0, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.0886305940149026, |
| "grad_norm": 14.359699249267578, |
| "learning_rate": 4.571348732050884e-06, |
| "loss": 0.3622, |
| "step": 14000 |
| }, |
| { |
| "epoch": 0.0886305940149026, |
| "eval_loss": 0.3693748712539673, |
| "eval_runtime": 115.6027, |
| "eval_samples_per_second": 1.73, |
| "eval_steps_per_second": 1.73, |
| "eval_wer": 100.0, |
| "step": 14000 |
| }, |
| { |
| "epoch": 0.10129210744560298, |
| "grad_norm": 34.34012222290039, |
| "learning_rate": 4.507840136162431e-06, |
| "loss": 0.3518, |
| "step": 16000 |
| }, |
| { |
| "epoch": 0.10129210744560298, |
| "eval_loss": 0.3601730465888977, |
| "eval_runtime": 100.1679, |
| "eval_samples_per_second": 1.997, |
| "eval_steps_per_second": 1.997, |
| "eval_wer": 100.0, |
| "step": 16000 |
| }, |
| { |
| "epoch": 0.11395362087630334, |
| "grad_norm": 13.935030937194824, |
| "learning_rate": 4.444331540273976e-06, |
| "loss": 0.3428, |
| "step": 18000 |
| }, |
| { |
| "epoch": 0.11395362087630334, |
| "eval_loss": 0.349777489900589, |
| "eval_runtime": 238.5911, |
| "eval_samples_per_second": 0.838, |
| "eval_steps_per_second": 0.838, |
| "eval_wer": 100.0, |
| "step": 18000 |
| }, |
| { |
| "epoch": 0.12661513430700372, |
| "grad_norm": 12.022929191589355, |
| "learning_rate": 4.380822944385523e-06, |
| "loss": 0.3322, |
| "step": 20000 |
| }, |
| { |
| "epoch": 0.12661513430700372, |
| "eval_loss": 0.3396839201450348, |
| "eval_runtime": 110.4827, |
| "eval_samples_per_second": 1.81, |
| "eval_steps_per_second": 1.81, |
| "eval_wer": 100.0, |
| "step": 20000 |
| }, |
| { |
| "epoch": 0.1392766477377041, |
| "grad_norm": 31.897066116333008, |
| "learning_rate": 4.3173143484970695e-06, |
| "loss": 0.3243, |
| "step": 22000 |
| }, |
| { |
| "epoch": 0.1392766477377041, |
| "eval_loss": 0.33299878239631653, |
| "eval_runtime": 112.72, |
| "eval_samples_per_second": 1.774, |
| "eval_steps_per_second": 1.774, |
| "eval_wer": 100.0, |
| "step": 22000 |
| }, |
| { |
| "epoch": 0.15193816116840447, |
| "grad_norm": 17.207157135009766, |
| "learning_rate": 4.2538057526086156e-06, |
| "loss": 0.3207, |
| "step": 24000 |
| }, |
| { |
| "epoch": 0.15193816116840447, |
| "eval_loss": 0.33643776178359985, |
| "eval_runtime": 99.1207, |
| "eval_samples_per_second": 2.018, |
| "eval_steps_per_second": 2.018, |
| "eval_wer": 100.0, |
| "step": 24000 |
| }, |
| { |
| "epoch": 0.16459967459910482, |
| "grad_norm": 29.34501075744629, |
| "learning_rate": 4.1902971567201624e-06, |
| "loss": 0.3118, |
| "step": 26000 |
| }, |
| { |
| "epoch": 0.16459967459910482, |
| "eval_loss": 0.3242020308971405, |
| "eval_runtime": 145.806, |
| "eval_samples_per_second": 1.372, |
| "eval_steps_per_second": 1.372, |
| "eval_wer": 100.0, |
| "step": 26000 |
| }, |
| { |
| "epoch": 0.1772611880298052, |
| "grad_norm": 13.88232421875, |
| "learning_rate": 4.1267885608317085e-06, |
| "loss": 0.3066, |
| "step": 28000 |
| }, |
| { |
| "epoch": 0.1772611880298052, |
| "eval_loss": 0.3177802264690399, |
| "eval_runtime": 117.2496, |
| "eval_samples_per_second": 1.706, |
| "eval_steps_per_second": 1.706, |
| "eval_wer": 100.0, |
| "step": 28000 |
| }, |
| { |
| "epoch": 0.18992270146050558, |
| "grad_norm": 10.003994941711426, |
| "learning_rate": 4.063279964943255e-06, |
| "loss": 0.3064, |
| "step": 30000 |
| }, |
| { |
| "epoch": 0.18992270146050558, |
| "eval_loss": 0.31194397807121277, |
| "eval_runtime": 98.5141, |
| "eval_samples_per_second": 2.03, |
| "eval_steps_per_second": 2.03, |
| "eval_wer": 100.0, |
| "step": 30000 |
| }, |
| { |
| "epoch": 0.20258421489120595, |
| "grad_norm": 27.381898880004883, |
| "learning_rate": 3.999771369054802e-06, |
| "loss": 0.3037, |
| "step": 32000 |
| }, |
| { |
| "epoch": 0.20258421489120595, |
| "eval_loss": 0.31087130308151245, |
| "eval_runtime": 115.818, |
| "eval_samples_per_second": 1.727, |
| "eval_steps_per_second": 1.727, |
| "eval_wer": 100.0, |
| "step": 32000 |
| }, |
| { |
| "epoch": 0.2152457283219063, |
| "grad_norm": 15.65439224243164, |
| "learning_rate": 3.936262773166348e-06, |
| "loss": 0.3001, |
| "step": 34000 |
| }, |
| { |
| "epoch": 0.2152457283219063, |
| "eval_loss": 0.30235540866851807, |
| "eval_runtime": 98.8755, |
| "eval_samples_per_second": 2.023, |
| "eval_steps_per_second": 2.023, |
| "eval_wer": 100.0, |
| "step": 34000 |
| }, |
| { |
| "epoch": 0.22790724175260668, |
| "grad_norm": 0.49147793650627136, |
| "learning_rate": 3.872754177277895e-06, |
| "loss": 0.2892, |
| "step": 36000 |
| }, |
| { |
| "epoch": 0.22790724175260668, |
| "eval_loss": 0.3077136278152466, |
| "eval_runtime": 167.6337, |
| "eval_samples_per_second": 1.193, |
| "eval_steps_per_second": 1.193, |
| "eval_wer": 100.0, |
| "step": 36000 |
| }, |
| { |
| "epoch": 0.24056875518330706, |
| "grad_norm": 23.69437026977539, |
| "learning_rate": 3.809245581389441e-06, |
| "loss": 0.2868, |
| "step": 38000 |
| }, |
| { |
| "epoch": 0.24056875518330706, |
| "eval_loss": 0.3034687936306, |
| "eval_runtime": 118.2635, |
| "eval_samples_per_second": 1.691, |
| "eval_steps_per_second": 1.691, |
| "eval_wer": 100.0, |
| "step": 38000 |
| }, |
| { |
| "epoch": 0.25323026861400744, |
| "grad_norm": 9.55330753326416, |
| "learning_rate": 3.745736985500988e-06, |
| "loss": 0.2869, |
| "step": 40000 |
| }, |
| { |
| "epoch": 0.25323026861400744, |
| "eval_loss": 0.3006068170070648, |
| "eval_runtime": 98.8792, |
| "eval_samples_per_second": 2.023, |
| "eval_steps_per_second": 2.023, |
| "eval_wer": 100.0, |
| "step": 40000 |
| }, |
| { |
| "epoch": 0.2658917820447078, |
| "grad_norm": 5.430517673492432, |
| "learning_rate": 3.6822283896125345e-06, |
| "loss": 0.3016, |
| "step": 42000 |
| }, |
| { |
| "epoch": 0.2658917820447078, |
| "eval_loss": 0.3174864649772644, |
| "eval_runtime": 120.0095, |
| "eval_samples_per_second": 1.667, |
| "eval_steps_per_second": 1.667, |
| "eval_wer": 100.0, |
| "step": 42000 |
| }, |
| { |
| "epoch": 0.2785532954754082, |
| "grad_norm": 2.0785741806030273, |
| "learning_rate": 3.618719793724081e-06, |
| "loss": 0.2745, |
| "step": 44000 |
| }, |
| { |
| "epoch": 0.2785532954754082, |
| "eval_loss": 0.2935592830181122, |
| "eval_runtime": 104.1383, |
| "eval_samples_per_second": 1.921, |
| "eval_steps_per_second": 1.921, |
| "eval_wer": 100.0, |
| "step": 44000 |
| }, |
| { |
| "epoch": 0.29121480890610857, |
| "grad_norm": 19.034826278686523, |
| "learning_rate": 3.5552111978356274e-06, |
| "loss": 0.2766, |
| "step": 46000 |
| }, |
| { |
| "epoch": 0.29121480890610857, |
| "eval_loss": 0.28297555446624756, |
| "eval_runtime": 128.9649, |
| "eval_samples_per_second": 1.551, |
| "eval_steps_per_second": 1.551, |
| "eval_wer": 100.0, |
| "step": 46000 |
| }, |
| { |
| "epoch": 0.30387632233680895, |
| "grad_norm": 5.441608905792236, |
| "learning_rate": 3.491702601947174e-06, |
| "loss": 0.2826, |
| "step": 48000 |
| }, |
| { |
| "epoch": 0.30387632233680895, |
| "eval_loss": 0.2824722230434418, |
| "eval_runtime": 111.7032, |
| "eval_samples_per_second": 1.79, |
| "eval_steps_per_second": 1.79, |
| "eval_wer": 100.0, |
| "step": 48000 |
| }, |
| { |
| "epoch": 0.31653783576750927, |
| "grad_norm": 25.043174743652344, |
| "learning_rate": 3.4281940060587203e-06, |
| "loss": 0.272, |
| "step": 50000 |
| }, |
| { |
| "epoch": 0.31653783576750927, |
| "eval_loss": 0.2822350263595581, |
| "eval_runtime": 99.249, |
| "eval_samples_per_second": 2.015, |
| "eval_steps_per_second": 2.015, |
| "eval_wer": 100.0, |
| "step": 50000 |
| }, |
| { |
| "epoch": 0.32919934919820965, |
| "grad_norm": 10.248320579528809, |
| "learning_rate": 3.364685410170267e-06, |
| "loss": 0.2721, |
| "step": 52000 |
| }, |
| { |
| "epoch": 0.32919934919820965, |
| "eval_loss": 0.2788018584251404, |
| "eval_runtime": 167.7003, |
| "eval_samples_per_second": 1.193, |
| "eval_steps_per_second": 1.193, |
| "eval_wer": 100.0, |
| "step": 52000 |
| }, |
| { |
| "epoch": 0.34186086262891, |
| "grad_norm": 15.632673263549805, |
| "learning_rate": 3.3011768142818136e-06, |
| "loss": 0.2748, |
| "step": 54000 |
| }, |
| { |
| "epoch": 0.34186086262891, |
| "eval_loss": 0.2818252146244049, |
| "eval_runtime": 115.5485, |
| "eval_samples_per_second": 1.731, |
| "eval_steps_per_second": 1.731, |
| "eval_wer": 100.0, |
| "step": 54000 |
| }, |
| { |
| "epoch": 0.3545223760596104, |
| "grad_norm": 22.526004791259766, |
| "learning_rate": 3.23766821839336e-06, |
| "loss": 0.2635, |
| "step": 56000 |
| }, |
| { |
| "epoch": 0.3545223760596104, |
| "eval_loss": 0.28214341402053833, |
| "eval_runtime": 98.4125, |
| "eval_samples_per_second": 2.032, |
| "eval_steps_per_second": 2.032, |
| "eval_wer": 100.0, |
| "step": 56000 |
| }, |
| { |
| "epoch": 0.3671838894903108, |
| "grad_norm": 13.596991539001465, |
| "learning_rate": 3.1741596225049065e-06, |
| "loss": 0.266, |
| "step": 58000 |
| }, |
| { |
| "epoch": 0.3671838894903108, |
| "eval_loss": 0.2786301374435425, |
| "eval_runtime": 112.9723, |
| "eval_samples_per_second": 1.77, |
| "eval_steps_per_second": 1.77, |
| "eval_wer": 100.0, |
| "step": 58000 |
| }, |
| { |
| "epoch": 0.37984540292101115, |
| "grad_norm": 20.539098739624023, |
| "learning_rate": 3.110651026616453e-06, |
| "loss": 0.2603, |
| "step": 60000 |
| }, |
| { |
| "epoch": 0.37984540292101115, |
| "eval_loss": 0.2836836278438568, |
| "eval_runtime": 98.7815, |
| "eval_samples_per_second": 2.025, |
| "eval_steps_per_second": 2.025, |
| "eval_wer": 100.0, |
| "step": 60000 |
| }, |
| { |
| "epoch": 0.39250691635171153, |
| "grad_norm": 9.462400436401367, |
| "learning_rate": 3.0471424307279994e-06, |
| "loss": 0.2589, |
| "step": 62000 |
| }, |
| { |
| "epoch": 0.39250691635171153, |
| "eval_loss": 0.28313395380973816, |
| "eval_runtime": 162.0004, |
| "eval_samples_per_second": 1.235, |
| "eval_steps_per_second": 1.235, |
| "eval_wer": 100.0, |
| "step": 62000 |
| }, |
| { |
| "epoch": 0.4051684297824119, |
| "grad_norm": 31.0218505859375, |
| "learning_rate": 2.9836338348395463e-06, |
| "loss": 0.2578, |
| "step": 64000 |
| }, |
| { |
| "epoch": 0.4051684297824119, |
| "eval_loss": 0.2877209782600403, |
| "eval_runtime": 119.6585, |
| "eval_samples_per_second": 1.671, |
| "eval_steps_per_second": 1.671, |
| "eval_wer": 100.0, |
| "step": 64000 |
| }, |
| { |
| "epoch": 0.4178299432131123, |
| "grad_norm": 16.72400665283203, |
| "learning_rate": 2.920125238951092e-06, |
| "loss": 0.2633, |
| "step": 66000 |
| }, |
| { |
| "epoch": 0.4178299432131123, |
| "eval_loss": 0.2801450490951538, |
| "eval_runtime": 102.711, |
| "eval_samples_per_second": 1.947, |
| "eval_steps_per_second": 1.947, |
| "eval_wer": 100.0, |
| "step": 66000 |
| }, |
| { |
| "epoch": 0.4304914566438126, |
| "grad_norm": 10.58651065826416, |
| "learning_rate": 2.8566166430626387e-06, |
| "loss": 0.2487, |
| "step": 68000 |
| }, |
| { |
| "epoch": 0.4304914566438126, |
| "eval_loss": 0.2773827016353607, |
| "eval_runtime": 119.5863, |
| "eval_samples_per_second": 1.672, |
| "eval_steps_per_second": 1.672, |
| "eval_wer": 100.0, |
| "step": 68000 |
| }, |
| { |
| "epoch": 0.443152970074513, |
| "grad_norm": 9.87272834777832, |
| "learning_rate": 2.793108047174185e-06, |
| "loss": 0.2429, |
| "step": 70000 |
| }, |
| { |
| "epoch": 0.443152970074513, |
| "eval_loss": 0.2784196734428406, |
| "eval_runtime": 110.2434, |
| "eval_samples_per_second": 1.814, |
| "eval_steps_per_second": 1.814, |
| "eval_wer": 100.0, |
| "step": 70000 |
| }, |
| { |
| "epoch": 0.45581448350521336, |
| "grad_norm": 17.528100967407227, |
| "learning_rate": 2.7295994512857316e-06, |
| "loss": 0.2474, |
| "step": 72000 |
| }, |
| { |
| "epoch": 0.45581448350521336, |
| "eval_loss": 0.275814414024353, |
| "eval_runtime": 99.8281, |
| "eval_samples_per_second": 2.003, |
| "eval_steps_per_second": 2.003, |
| "eval_wer": 100.0, |
| "step": 72000 |
| }, |
| { |
| "epoch": 0.46847599693591374, |
| "grad_norm": 26.67496681213379, |
| "learning_rate": 2.666090855397278e-06, |
| "loss": 0.2504, |
| "step": 74000 |
| }, |
| { |
| "epoch": 0.46847599693591374, |
| "eval_loss": 0.26936283707618713, |
| "eval_runtime": 175.1329, |
| "eval_samples_per_second": 1.142, |
| "eval_steps_per_second": 1.142, |
| "eval_wer": 100.0, |
| "step": 74000 |
| }, |
| { |
| "epoch": 0.4811375103666141, |
| "grad_norm": 7.970252990722656, |
| "learning_rate": 2.6025822595088245e-06, |
| "loss": 0.2571, |
| "step": 76000 |
| }, |
| { |
| "epoch": 0.4811375103666141, |
| "eval_loss": 0.26590612530708313, |
| "eval_runtime": 116.1576, |
| "eval_samples_per_second": 1.722, |
| "eval_steps_per_second": 1.722, |
| "eval_wer": 100.0, |
| "step": 76000 |
| }, |
| { |
| "epoch": 0.4937990237973145, |
| "grad_norm": 11.1566743850708, |
| "learning_rate": 2.539073663620371e-06, |
| "loss": 0.2483, |
| "step": 78000 |
| }, |
| { |
| "epoch": 0.4937990237973145, |
| "eval_loss": 0.2682250738143921, |
| "eval_runtime": 98.2738, |
| "eval_samples_per_second": 2.035, |
| "eval_steps_per_second": 2.035, |
| "eval_wer": 100.0, |
| "step": 78000 |
| }, |
| { |
| "epoch": 0.5064605372280149, |
| "grad_norm": 10.013092041015625, |
| "learning_rate": 2.4755650677319174e-06, |
| "loss": 0.245, |
| "step": 80000 |
| }, |
| { |
| "epoch": 0.5064605372280149, |
| "eval_loss": 0.2621174156665802, |
| "eval_runtime": 113.9651, |
| "eval_samples_per_second": 1.755, |
| "eval_steps_per_second": 1.755, |
| "eval_wer": 100.0, |
| "step": 80000 |
| }, |
| { |
| "epoch": 0.5191220506587152, |
| "grad_norm": 19.42994499206543, |
| "learning_rate": 2.4120564718434643e-06, |
| "loss": 0.2584, |
| "step": 82000 |
| }, |
| { |
| "epoch": 0.5191220506587152, |
| "eval_loss": 0.2639446556568146, |
| "eval_runtime": 100.2051, |
| "eval_samples_per_second": 1.996, |
| "eval_steps_per_second": 1.996, |
| "eval_wer": 100.0, |
| "step": 82000 |
| }, |
| { |
| "epoch": 0.5317835640894156, |
| "grad_norm": 4.659946918487549, |
| "learning_rate": 2.3485478759550108e-06, |
| "loss": 0.244, |
| "step": 84000 |
| }, |
| { |
| "epoch": 0.5317835640894156, |
| "eval_loss": 0.2655565142631531, |
| "eval_runtime": 139.2135, |
| "eval_samples_per_second": 1.437, |
| "eval_steps_per_second": 1.437, |
| "eval_wer": 100.0, |
| "step": 84000 |
| }, |
| { |
| "epoch": 0.544445077520116, |
| "grad_norm": 14.925030708312988, |
| "learning_rate": 2.2850392800665572e-06, |
| "loss": 0.2408, |
| "step": 86000 |
| }, |
| { |
| "epoch": 0.544445077520116, |
| "eval_loss": 0.2684570848941803, |
| "eval_runtime": 117.6776, |
| "eval_samples_per_second": 1.7, |
| "eval_steps_per_second": 1.7, |
| "eval_wer": 100.0, |
| "step": 86000 |
| }, |
| { |
| "epoch": 0.5571065909508164, |
| "grad_norm": 17.363605499267578, |
| "learning_rate": 2.2215306841781037e-06, |
| "loss": 0.2398, |
| "step": 88000 |
| }, |
| { |
| "epoch": 0.5571065909508164, |
| "eval_loss": 0.2665661573410034, |
| "eval_runtime": 100.8416, |
| "eval_samples_per_second": 1.983, |
| "eval_steps_per_second": 1.983, |
| "eval_wer": 100.0, |
| "step": 88000 |
| }, |
| { |
| "epoch": 0.5697681043815167, |
| "grad_norm": 7.914783477783203, |
| "learning_rate": 2.15802208828965e-06, |
| "loss": 0.2417, |
| "step": 90000 |
| }, |
| { |
| "epoch": 0.5697681043815167, |
| "eval_loss": 0.25880980491638184, |
| "eval_runtime": 114.0038, |
| "eval_samples_per_second": 1.754, |
| "eval_steps_per_second": 1.754, |
| "eval_wer": 100.0, |
| "step": 90000 |
| }, |
| { |
| "epoch": 0.5824296178122171, |
| "grad_norm": 6.1498517990112305, |
| "learning_rate": 2.0945134924011966e-06, |
| "loss": 0.2367, |
| "step": 92000 |
| }, |
| { |
| "epoch": 0.5824296178122171, |
| "eval_loss": 0.25671330094337463, |
| "eval_runtime": 99.2437, |
| "eval_samples_per_second": 2.015, |
| "eval_steps_per_second": 2.015, |
| "eval_wer": 100.0, |
| "step": 92000 |
| }, |
| { |
| "epoch": 0.5950911312429175, |
| "grad_norm": 15.99548625946045, |
| "learning_rate": 2.0310048965127434e-06, |
| "loss": 0.2492, |
| "step": 94000 |
| }, |
| { |
| "epoch": 0.5950911312429175, |
| "eval_loss": 0.256805419921875, |
| "eval_runtime": 157.1247, |
| "eval_samples_per_second": 1.273, |
| "eval_steps_per_second": 1.273, |
| "eval_wer": 100.0, |
| "step": 94000 |
| }, |
| { |
| "epoch": 0.6077526446736179, |
| "grad_norm": 24.078039169311523, |
| "learning_rate": 1.96749630062429e-06, |
| "loss": 0.234, |
| "step": 96000 |
| }, |
| { |
| "epoch": 0.6077526446736179, |
| "eval_loss": 0.25336185097694397, |
| "eval_runtime": 117.2279, |
| "eval_samples_per_second": 1.706, |
| "eval_steps_per_second": 1.706, |
| "eval_wer": 100.0, |
| "step": 96000 |
| }, |
| { |
| "epoch": 0.6204141581043182, |
| "grad_norm": 13.121466636657715, |
| "learning_rate": 1.9039877047358363e-06, |
| "loss": 0.2338, |
| "step": 98000 |
| }, |
| { |
| "epoch": 0.6204141581043182, |
| "eval_loss": 0.255278617143631, |
| "eval_runtime": 99.8788, |
| "eval_samples_per_second": 2.002, |
| "eval_steps_per_second": 2.002, |
| "eval_wer": 100.0, |
| "step": 98000 |
| }, |
| { |
| "epoch": 0.6330756715350185, |
| "grad_norm": 9.44789981842041, |
| "learning_rate": 1.8404791088473826e-06, |
| "loss": 0.2307, |
| "step": 100000 |
| }, |
| { |
| "epoch": 0.6330756715350185, |
| "eval_loss": 0.2485535740852356, |
| "eval_runtime": 114.476, |
| "eval_samples_per_second": 1.747, |
| "eval_steps_per_second": 1.747, |
| "eval_wer": 100.0, |
| "step": 100000 |
| }, |
| { |
| "epoch": 0.645737184965719, |
| "grad_norm": 16.697378158569336, |
| "learning_rate": 1.776970512958929e-06, |
| "loss": 0.2343, |
| "step": 102000 |
| }, |
| { |
| "epoch": 0.645737184965719, |
| "eval_loss": 0.2523353695869446, |
| "eval_runtime": 99.8529, |
| "eval_samples_per_second": 2.003, |
| "eval_steps_per_second": 2.003, |
| "eval_wer": 100.0, |
| "step": 102000 |
| }, |
| { |
| "epoch": 0.6583986983964193, |
| "grad_norm": 11.517345428466797, |
| "learning_rate": 1.7134619170704755e-06, |
| "loss": 0.2366, |
| "step": 104000 |
| }, |
| { |
| "epoch": 0.6583986983964193, |
| "eval_loss": 0.2554631531238556, |
| "eval_runtime": 135.1832, |
| "eval_samples_per_second": 1.479, |
| "eval_steps_per_second": 1.479, |
| "eval_wer": 100.0, |
| "step": 104000 |
| }, |
| { |
| "epoch": 0.6710602118271197, |
| "grad_norm": 22.604074478149414, |
| "learning_rate": 1.6499533211820221e-06, |
| "loss": 0.2323, |
| "step": 106000 |
| }, |
| { |
| "epoch": 0.6710602118271197, |
| "eval_loss": 0.2547050416469574, |
| "eval_runtime": 121.0754, |
| "eval_samples_per_second": 1.652, |
| "eval_steps_per_second": 1.652, |
| "eval_wer": 100.0, |
| "step": 106000 |
| }, |
| { |
| "epoch": 0.68372172525782, |
| "grad_norm": 14.323906898498535, |
| "learning_rate": 1.5864447252935686e-06, |
| "loss": 0.2343, |
| "step": 108000 |
| }, |
| { |
| "epoch": 0.68372172525782, |
| "eval_loss": 0.2524075508117676, |
| "eval_runtime": 103.5538, |
| "eval_samples_per_second": 1.931, |
| "eval_steps_per_second": 1.931, |
| "eval_wer": 100.0, |
| "step": 108000 |
| }, |
| { |
| "epoch": 0.6963832386885205, |
| "grad_norm": 11.585010528564453, |
| "learning_rate": 1.522936129405115e-06, |
| "loss": 0.2348, |
| "step": 110000 |
| }, |
| { |
| "epoch": 0.6963832386885205, |
| "eval_loss": 0.24542486667633057, |
| "eval_runtime": 124.354, |
| "eval_samples_per_second": 1.608, |
| "eval_steps_per_second": 1.608, |
| "eval_wer": 100.0, |
| "step": 110000 |
| }, |
| { |
| "epoch": 0.7090447521192208, |
| "grad_norm": 13.175851821899414, |
| "learning_rate": 1.4594275335166617e-06, |
| "loss": 0.245, |
| "step": 112000 |
| }, |
| { |
| "epoch": 0.7090447521192208, |
| "eval_loss": 0.24049542844295502, |
| "eval_runtime": 112.7155, |
| "eval_samples_per_second": 1.774, |
| "eval_steps_per_second": 1.774, |
| "eval_wer": 100.0, |
| "step": 112000 |
| }, |
| { |
| "epoch": 0.7217062655499212, |
| "grad_norm": 5.587381839752197, |
| "learning_rate": 1.3959189376282082e-06, |
| "loss": 0.2332, |
| "step": 114000 |
| }, |
| { |
| "epoch": 0.7217062655499212, |
| "eval_loss": 0.23937122523784637, |
| "eval_runtime": 102.231, |
| "eval_samples_per_second": 1.956, |
| "eval_steps_per_second": 1.956, |
| "eval_wer": 100.0, |
| "step": 114000 |
| }, |
| { |
| "epoch": 0.7343677789806216, |
| "grad_norm": 39.2594108581543, |
| "learning_rate": 1.3324103417397546e-06, |
| "loss": 0.2251, |
| "step": 116000 |
| }, |
| { |
| "epoch": 0.7343677789806216, |
| "eval_loss": 0.24494153261184692, |
| "eval_runtime": 111.2917, |
| "eval_samples_per_second": 1.797, |
| "eval_steps_per_second": 1.797, |
| "eval_wer": 100.0, |
| "step": 116000 |
| }, |
| { |
| "epoch": 0.7470292924113219, |
| "grad_norm": 13.437777519226074, |
| "learning_rate": 1.268901745851301e-06, |
| "loss": 0.2279, |
| "step": 118000 |
| }, |
| { |
| "epoch": 0.7470292924113219, |
| "eval_loss": 0.24640831351280212, |
| "eval_runtime": 100.079, |
| "eval_samples_per_second": 1.998, |
| "eval_steps_per_second": 1.998, |
| "eval_wer": 100.0, |
| "step": 118000 |
| }, |
| { |
| "epoch": 0.7596908058420223, |
| "grad_norm": 17.973445892333984, |
| "learning_rate": 1.2053931499628475e-06, |
| "loss": 0.2269, |
| "step": 120000 |
| }, |
| { |
| "epoch": 0.7596908058420223, |
| "eval_loss": 0.2451734095811844, |
| "eval_runtime": 150.8961, |
| "eval_samples_per_second": 1.325, |
| "eval_steps_per_second": 1.325, |
| "eval_wer": 100.0, |
| "step": 120000 |
| }, |
| { |
| "epoch": 0.7723523192727226, |
| "grad_norm": 1.6845195293426514, |
| "learning_rate": 1.141884554074394e-06, |
| "loss": 0.2303, |
| "step": 122000 |
| }, |
| { |
| "epoch": 0.7723523192727226, |
| "eval_loss": 0.23949329555034637, |
| "eval_runtime": 119.1671, |
| "eval_samples_per_second": 1.678, |
| "eval_steps_per_second": 1.678, |
| "eval_wer": 100.0, |
| "step": 122000 |
| }, |
| { |
| "epoch": 0.7850138327034231, |
| "grad_norm": 19.897830963134766, |
| "learning_rate": 1.0783759581859406e-06, |
| "loss": 0.2127, |
| "step": 124000 |
| }, |
| { |
| "epoch": 0.7850138327034231, |
| "eval_loss": 0.24049174785614014, |
| "eval_runtime": 102.2185, |
| "eval_samples_per_second": 1.957, |
| "eval_steps_per_second": 1.957, |
| "eval_wer": 100.0, |
| "step": 124000 |
| }, |
| { |
| "epoch": 0.7976753461341234, |
| "grad_norm": 12.614727973937988, |
| "learning_rate": 1.014867362297487e-06, |
| "loss": 0.2289, |
| "step": 126000 |
| }, |
| { |
| "epoch": 0.7976753461341234, |
| "eval_loss": 0.2411843240261078, |
| "eval_runtime": 116.078, |
| "eval_samples_per_second": 1.723, |
| "eval_steps_per_second": 1.723, |
| "eval_wer": 100.0, |
| "step": 126000 |
| }, |
| { |
| "epoch": 0.8103368595648238, |
| "grad_norm": 19.16327476501465, |
| "learning_rate": 9.513587664090336e-07, |
| "loss": 0.2169, |
| "step": 128000 |
| }, |
| { |
| "epoch": 0.8103368595648238, |
| "eval_loss": 0.2375524491071701, |
| "eval_runtime": 99.6286, |
| "eval_samples_per_second": 2.007, |
| "eval_steps_per_second": 2.007, |
| "eval_wer": 100.0, |
| "step": 128000 |
| }, |
| { |
| "epoch": 0.8229983729955241, |
| "grad_norm": 17.00189781188965, |
| "learning_rate": 8.8785017052058e-07, |
| "loss": 0.234, |
| "step": 130000 |
| }, |
| { |
| "epoch": 0.8229983729955241, |
| "eval_loss": 0.23726019263267517, |
| "eval_runtime": 128.4372, |
| "eval_samples_per_second": 1.557, |
| "eval_steps_per_second": 1.557, |
| "eval_wer": 100.0, |
| "step": 130000 |
| }, |
| { |
| "epoch": 0.8356598864262246, |
| "grad_norm": 7.840052127838135, |
| "learning_rate": 8.243415746321265e-07, |
| "loss": 0.2282, |
| "step": 132000 |
| }, |
| { |
| "epoch": 0.8356598864262246, |
| "eval_loss": 0.23711390793323517, |
| "eval_runtime": 118.2074, |
| "eval_samples_per_second": 1.692, |
| "eval_steps_per_second": 1.692, |
| "eval_wer": 100.0, |
| "step": 132000 |
| }, |
| { |
| "epoch": 0.8483213998569249, |
| "grad_norm": 23.74713706970215, |
| "learning_rate": 7.60832978743673e-07, |
| "loss": 0.2323, |
| "step": 134000 |
| }, |
| { |
| "epoch": 0.8483213998569249, |
| "eval_loss": 0.232102170586586, |
| "eval_runtime": 102.3613, |
| "eval_samples_per_second": 1.954, |
| "eval_steps_per_second": 1.954, |
| "eval_wer": 100.0, |
| "step": 134000 |
| }, |
| { |
| "epoch": 0.8609829132876252, |
| "grad_norm": 10.603443145751953, |
| "learning_rate": 6.973243828552195e-07, |
| "loss": 0.2222, |
| "step": 136000 |
| }, |
| { |
| "epoch": 0.8609829132876252, |
| "eval_loss": 0.23403330147266388, |
| "eval_runtime": 122.2087, |
| "eval_samples_per_second": 1.637, |
| "eval_steps_per_second": 1.637, |
| "eval_wer": 100.0, |
| "step": 136000 |
| }, |
| { |
| "epoch": 0.8736444267183257, |
| "grad_norm": 14.557578086853027, |
| "learning_rate": 6.338157869667661e-07, |
| "loss": 0.2223, |
| "step": 138000 |
| }, |
| { |
| "epoch": 0.8736444267183257, |
| "eval_loss": 0.23369312286376953, |
| "eval_runtime": 111.0931, |
| "eval_samples_per_second": 1.8, |
| "eval_steps_per_second": 1.8, |
| "eval_wer": 100.0, |
| "step": 138000 |
| }, |
| { |
| "epoch": 0.886305940149026, |
| "grad_norm": 21.674922943115234, |
| "learning_rate": 5.703071910783125e-07, |
| "loss": 0.2211, |
| "step": 140000 |
| }, |
| { |
| "epoch": 0.886305940149026, |
| "eval_loss": 0.2323533594608307, |
| "eval_runtime": 103.9127, |
| "eval_samples_per_second": 1.925, |
| "eval_steps_per_second": 1.925, |
| "eval_wer": 100.0, |
| "step": 140000 |
| }, |
| { |
| "epoch": 0.8989674535797264, |
| "grad_norm": 4.280592918395996, |
| "learning_rate": 5.06798595189859e-07, |
| "loss": 0.2066, |
| "step": 142000 |
| }, |
| { |
| "epoch": 0.8989674535797264, |
| "eval_loss": 0.23401407897472382, |
| "eval_runtime": 112.7995, |
| "eval_samples_per_second": 1.773, |
| "eval_steps_per_second": 1.773, |
| "eval_wer": 100.0, |
| "step": 142000 |
| }, |
| { |
| "epoch": 0.9116289670104267, |
| "grad_norm": 3.850653886795044, |
| "learning_rate": 4.4328999930140544e-07, |
| "loss": 0.2283, |
| "step": 144000 |
| }, |
| { |
| "epoch": 0.9116289670104267, |
| "eval_loss": 0.2337852418422699, |
| "eval_runtime": 102.0048, |
| "eval_samples_per_second": 1.961, |
| "eval_steps_per_second": 1.961, |
| "eval_wer": 100.0, |
| "step": 144000 |
| }, |
| { |
| "epoch": 0.9242904804411272, |
| "grad_norm": 7.110324382781982, |
| "learning_rate": 3.7978140341295195e-07, |
| "loss": 0.2232, |
| "step": 146000 |
| }, |
| { |
| "epoch": 0.9242904804411272, |
| "eval_loss": 0.23404987156391144, |
| "eval_runtime": 118.1946, |
| "eval_samples_per_second": 1.692, |
| "eval_steps_per_second": 1.692, |
| "eval_wer": 100.0, |
| "step": 146000 |
| }, |
| { |
| "epoch": 0.9369519938718275, |
| "grad_norm": 4.778629302978516, |
| "learning_rate": 3.162728075244985e-07, |
| "loss": 0.2184, |
| "step": 148000 |
| }, |
| { |
| "epoch": 0.9369519938718275, |
| "eval_loss": 0.2326117902994156, |
| "eval_runtime": 118.5384, |
| "eval_samples_per_second": 1.687, |
| "eval_steps_per_second": 1.687, |
| "eval_wer": 100.0, |
| "step": 148000 |
| }, |
| { |
| "epoch": 0.9496135073025279, |
| "grad_norm": 7.637845039367676, |
| "learning_rate": 2.5276421163604495e-07, |
| "loss": 0.218, |
| "step": 150000 |
| }, |
| { |
| "epoch": 0.9496135073025279, |
| "eval_loss": 0.2330457717180252, |
| "eval_runtime": 103.0085, |
| "eval_samples_per_second": 1.942, |
| "eval_steps_per_second": 1.942, |
| "eval_wer": 100.0, |
| "step": 150000 |
| }, |
| { |
| "epoch": 0.9622750207332282, |
| "grad_norm": 18.473331451416016, |
| "learning_rate": 1.8925561574759146e-07, |
| "loss": 0.2217, |
| "step": 152000 |
| }, |
| { |
| "epoch": 0.9622750207332282, |
| "eval_loss": 0.23135539889335632, |
| "eval_runtime": 116.9612, |
| "eval_samples_per_second": 1.71, |
| "eval_steps_per_second": 1.71, |
| "eval_wer": 100.0, |
| "step": 152000 |
| }, |
| { |
| "epoch": 0.9749365341639287, |
| "grad_norm": 9.762419700622559, |
| "learning_rate": 1.2574701985913796e-07, |
| "loss": 0.2197, |
| "step": 154000 |
| }, |
| { |
| "epoch": 0.9749365341639287, |
| "eval_loss": 0.23144523799419403, |
| "eval_runtime": 101.6277, |
| "eval_samples_per_second": 1.968, |
| "eval_steps_per_second": 1.968, |
| "eval_wer": 100.0, |
| "step": 154000 |
| }, |
| { |
| "epoch": 0.987598047594629, |
| "grad_norm": 7.513221263885498, |
| "learning_rate": 6.223842397068444e-08, |
| "loss": 0.2157, |
| "step": 156000 |
| }, |
| { |
| "epoch": 0.987598047594629, |
| "eval_loss": 0.23098862171173096, |
| "eval_runtime": 110.3102, |
| "eval_samples_per_second": 1.813, |
| "eval_steps_per_second": 1.813, |
| "eval_wer": 100.0, |
| "step": 156000 |
| } |
| ], |
| "logging_steps": 2000, |
| "max_steps": 157959, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 2000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 9.116903391289344e+19, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
