| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.536067892503536, | |
| "eval_steps": 100, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14144271570014144, | |
| "eval_loss": 4.035430908203125, | |
| "eval_runtime": 153.5895, | |
| "eval_samples_per_second": 36.825, | |
| "eval_steps_per_second": 4.603, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2828854314002829, | |
| "eval_loss": 3.097731590270996, | |
| "eval_runtime": 149.7934, | |
| "eval_samples_per_second": 37.759, | |
| "eval_steps_per_second": 4.72, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4243281471004243, | |
| "eval_loss": 3.0768725872039795, | |
| "eval_runtime": 151.0662, | |
| "eval_samples_per_second": 37.441, | |
| "eval_steps_per_second": 4.68, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5657708628005658, | |
| "eval_loss": 1.373844027519226, | |
| "eval_runtime": 150.8632, | |
| "eval_samples_per_second": 37.491, | |
| "eval_steps_per_second": 4.686, | |
| "eval_wer": 0.8913622648777043, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7072135785007072, | |
| "grad_norm": 2.2486917972564697, | |
| "learning_rate": 0.0002964, | |
| "loss": 3.7586, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7072135785007072, | |
| "eval_loss": 1.0915191173553467, | |
| "eval_runtime": 151.7555, | |
| "eval_samples_per_second": 37.27, | |
| "eval_steps_per_second": 4.659, | |
| "eval_wer": 0.7692431148488156, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8486562942008486, | |
| "eval_loss": 0.9360549449920654, | |
| "eval_runtime": 152.9408, | |
| "eval_samples_per_second": 36.982, | |
| "eval_steps_per_second": 4.623, | |
| "eval_wer": 0.6854818000898761, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9900990099009901, | |
| "eval_loss": 0.8494995832443237, | |
| "eval_runtime": 151.5832, | |
| "eval_samples_per_second": 37.313, | |
| "eval_steps_per_second": 4.664, | |
| "eval_wer": 0.6247191371894459, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1315417256011315, | |
| "eval_loss": 0.6885886788368225, | |
| "eval_runtime": 151.913, | |
| "eval_samples_per_second": 37.232, | |
| "eval_steps_per_second": 4.654, | |
| "eval_wer": 0.5397059767606086, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.272984441301273, | |
| "eval_loss": 0.6703779101371765, | |
| "eval_runtime": 153.2384, | |
| "eval_samples_per_second": 36.91, | |
| "eval_steps_per_second": 4.614, | |
| "eval_wer": 0.531215895230147, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4144271570014144, | |
| "grad_norm": 0.8719882369041443, | |
| "learning_rate": 0.0002259, | |
| "loss": 0.8877, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.4144271570014144, | |
| "eval_loss": 0.62369304895401, | |
| "eval_runtime": 153.3779, | |
| "eval_samples_per_second": 36.876, | |
| "eval_steps_per_second": 4.61, | |
| "eval_wer": 0.495040765230789, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5558698727015559, | |
| "eval_loss": 0.5992260575294495, | |
| "eval_runtime": 152.5492, | |
| "eval_samples_per_second": 37.077, | |
| "eval_steps_per_second": 4.635, | |
| "eval_wer": 0.4767606085895872, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6973125884016973, | |
| "eval_loss": 0.5729629397392273, | |
| "eval_runtime": 152.035, | |
| "eval_samples_per_second": 37.202, | |
| "eval_steps_per_second": 4.65, | |
| "eval_wer": 0.4521730756885151, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.8387553041018387, | |
| "eval_loss": 0.5504249930381775, | |
| "eval_runtime": 152.762, | |
| "eval_samples_per_second": 37.025, | |
| "eval_steps_per_second": 4.628, | |
| "eval_wer": 0.4417891763497464, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.9801980198019802, | |
| "eval_loss": 0.5288310050964355, | |
| "eval_runtime": 152.1876, | |
| "eval_samples_per_second": 37.165, | |
| "eval_steps_per_second": 4.646, | |
| "eval_wer": 0.42586826731719846, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.1216407355021216, | |
| "grad_norm": 0.7933566570281982, | |
| "learning_rate": 0.00015134999999999997, | |
| "loss": 0.6844, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.1216407355021216, | |
| "eval_loss": 0.5165240168571472, | |
| "eval_runtime": 152.8573, | |
| "eval_samples_per_second": 37.002, | |
| "eval_steps_per_second": 4.625, | |
| "eval_wer": 0.4216633498106182, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.263083451202263, | |
| "eval_loss": 0.5071681141853333, | |
| "eval_runtime": 152.7102, | |
| "eval_samples_per_second": 37.037, | |
| "eval_steps_per_second": 4.63, | |
| "eval_wer": 0.41930410220196446, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.4045261669024045, | |
| "eval_loss": 0.49842193722724915, | |
| "eval_runtime": 153.4952, | |
| "eval_samples_per_second": 36.848, | |
| "eval_steps_per_second": 4.606, | |
| "eval_wer": 0.41546831867496953, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.545968882602546, | |
| "eval_loss": 0.4882013499736786, | |
| "eval_runtime": 152.4926, | |
| "eval_samples_per_second": 37.09, | |
| "eval_steps_per_second": 4.636, | |
| "eval_wer": 0.4096905694292868, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.6874115983026874, | |
| "eval_loss": 0.4803565740585327, | |
| "eval_runtime": 152.0827, | |
| "eval_samples_per_second": 37.19, | |
| "eval_steps_per_second": 4.649, | |
| "eval_wer": 0.40800539256596263, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.828854314002829, | |
| "grad_norm": 0.7038583159446716, | |
| "learning_rate": 7.68e-05, | |
| "loss": 0.537, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.828854314002829, | |
| "eval_loss": 0.470017671585083, | |
| "eval_runtime": 152.5003, | |
| "eval_samples_per_second": 37.088, | |
| "eval_steps_per_second": 4.636, | |
| "eval_wer": 0.39272645567182385, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.9702970297029703, | |
| "eval_loss": 0.46773043274879456, | |
| "eval_runtime": 152.3208, | |
| "eval_samples_per_second": 37.132, | |
| "eval_steps_per_second": 4.642, | |
| "eval_wer": 0.3884573409514027, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.1117397454031117, | |
| "eval_loss": 0.4683248996734619, | |
| "eval_runtime": 152.8021, | |
| "eval_samples_per_second": 37.015, | |
| "eval_steps_per_second": 4.627, | |
| "eval_wer": 0.38572895936316365, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.253182461103253, | |
| "eval_loss": 0.4618384838104248, | |
| "eval_runtime": 152.4829, | |
| "eval_samples_per_second": 37.093, | |
| "eval_steps_per_second": 4.637, | |
| "eval_wer": 0.3792289914617706, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.3946251768033946, | |
| "eval_loss": 0.46039119362831116, | |
| "eval_runtime": 152.4542, | |
| "eval_samples_per_second": 37.1, | |
| "eval_steps_per_second": 4.637, | |
| "eval_wer": 0.3762759196250883, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.536067892503536, | |
| "grad_norm": 0.6644078493118286, | |
| "learning_rate": 2.1e-06, | |
| "loss": 0.4434, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.536067892503536, | |
| "eval_loss": 0.4588949382305145, | |
| "eval_runtime": 152.8982, | |
| "eval_samples_per_second": 36.992, | |
| "eval_steps_per_second": 4.624, | |
| "eval_wer": 0.3742697566925595, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.536067892503536, | |
| "step": 2500, | |
| "total_flos": 9.55169606524761e+18, | |
| "train_loss": 1.2622116821289062, | |
| "train_runtime": 6522.8506, | |
| "train_samples_per_second": 12.265, | |
| "train_steps_per_second": 0.383 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 400, | |
| "total_flos": 9.55169606524761e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |