| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 30.864197530864196, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0625, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_loss": 0.28051668405532837, | |
| "eval_runtime": 198.6071, | |
| "eval_samples_per_second": 16.374, | |
| "eval_steps_per_second": 2.049, | |
| "eval_wer": 0.2381097005406062, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 9.987639060568604e-05, | |
| "loss": 0.047, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_loss": 0.34423157572746277, | |
| "eval_runtime": 147.7154, | |
| "eval_samples_per_second": 22.015, | |
| "eval_steps_per_second": 2.755, | |
| "eval_wer": 0.2334325457085586, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 9.975278121137207e-05, | |
| "loss": 0.0409, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_loss": 0.3597787916660309, | |
| "eval_runtime": 156.092, | |
| "eval_samples_per_second": 20.834, | |
| "eval_steps_per_second": 2.607, | |
| "eval_wer": 0.23124582396889995, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 9.96291718170581e-05, | |
| "loss": 0.0413, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_loss": 0.35860675573349, | |
| "eval_runtime": 151.5312, | |
| "eval_samples_per_second": 21.461, | |
| "eval_steps_per_second": 2.686, | |
| "eval_wer": 0.2421794326671931, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 9.950556242274414e-05, | |
| "loss": 0.0388, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "eval_loss": 0.3748931586742401, | |
| "eval_runtime": 154.1563, | |
| "eval_samples_per_second": 21.095, | |
| "eval_steps_per_second": 2.64, | |
| "eval_wer": 0.23914231913988945, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 9.938195302843017e-05, | |
| "loss": 0.0383, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "eval_loss": 0.36109668016433716, | |
| "eval_runtime": 163.2656, | |
| "eval_samples_per_second": 19.918, | |
| "eval_steps_per_second": 2.493, | |
| "eval_wer": 0.2352548138249408, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "learning_rate": 9.92583436341162e-05, | |
| "loss": 0.0381, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "eval_loss": 0.3883003294467926, | |
| "eval_runtime": 164.0468, | |
| "eval_samples_per_second": 19.824, | |
| "eval_steps_per_second": 2.481, | |
| "eval_wer": 0.23683411285913866, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "learning_rate": 9.913473423980223e-05, | |
| "loss": 0.0379, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "eval_loss": 0.3676028549671173, | |
| "eval_runtime": 174.6686, | |
| "eval_samples_per_second": 18.618, | |
| "eval_steps_per_second": 2.33, | |
| "eval_wer": 0.2324606693798214, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 9.901112484548825e-05, | |
| "loss": 0.0364, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "eval_loss": 0.37265580892562866, | |
| "eval_runtime": 150.5621, | |
| "eval_samples_per_second": 21.599, | |
| "eval_steps_per_second": 2.703, | |
| "eval_wer": 0.23331106116746644, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 12.35, | |
| "learning_rate": 9.88875154511743e-05, | |
| "loss": 0.0355, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 12.35, | |
| "eval_loss": 0.3740839958190918, | |
| "eval_runtime": 180.5633, | |
| "eval_samples_per_second": 18.01, | |
| "eval_steps_per_second": 2.254, | |
| "eval_wer": 0.23282512300309785, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.58, | |
| "learning_rate": 9.876390605686032e-05, | |
| "loss": 0.0365, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 13.58, | |
| "eval_loss": 0.37167179584503174, | |
| "eval_runtime": 177.8674, | |
| "eval_samples_per_second": 18.283, | |
| "eval_steps_per_second": 2.288, | |
| "eval_wer": 0.23580149425985544, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 14.81, | |
| "learning_rate": 9.864029666254637e-05, | |
| "loss": 0.0343, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 14.81, | |
| "eval_loss": 0.3842860460281372, | |
| "eval_runtime": 182.4399, | |
| "eval_samples_per_second": 17.825, | |
| "eval_steps_per_second": 2.231, | |
| "eval_wer": 0.23580149425985544, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 16.05, | |
| "learning_rate": 9.851668726823239e-05, | |
| "loss": 0.0358, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 16.05, | |
| "eval_loss": 0.3594246208667755, | |
| "eval_runtime": 180.3747, | |
| "eval_samples_per_second": 18.029, | |
| "eval_steps_per_second": 2.256, | |
| "eval_wer": 0.23616594788313186, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 17.28, | |
| "learning_rate": 9.839307787391843e-05, | |
| "loss": 0.0343, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 17.28, | |
| "eval_loss": 0.4051465690135956, | |
| "eval_runtime": 180.0644, | |
| "eval_samples_per_second": 18.06, | |
| "eval_steps_per_second": 2.26, | |
| "eval_wer": 0.2351940715543947, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "learning_rate": 9.826946847960445e-05, | |
| "loss": 0.0363, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "eval_loss": 0.3952256739139557, | |
| "eval_runtime": 185.3946, | |
| "eval_samples_per_second": 17.541, | |
| "eval_steps_per_second": 2.195, | |
| "eval_wer": 0.23962825730425805, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 19.75, | |
| "learning_rate": 9.814585908529048e-05, | |
| "loss": 0.0382, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 19.75, | |
| "eval_loss": 0.3582073450088501, | |
| "eval_runtime": 179.8373, | |
| "eval_samples_per_second": 18.083, | |
| "eval_steps_per_second": 2.263, | |
| "eval_wer": 0.23756302010569155, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 20.99, | |
| "learning_rate": 9.802224969097652e-05, | |
| "loss": 0.0337, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 20.99, | |
| "eval_loss": 0.3877179026603699, | |
| "eval_runtime": 183.3138, | |
| "eval_samples_per_second": 17.74, | |
| "eval_steps_per_second": 2.22, | |
| "eval_wer": 0.2347081333900261, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "learning_rate": 9.789864029666255e-05, | |
| "loss": 0.0331, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "eval_loss": 0.3826364576816559, | |
| "eval_runtime": 174.2395, | |
| "eval_samples_per_second": 18.664, | |
| "eval_steps_per_second": 2.336, | |
| "eval_wer": 0.2383526696227905, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 23.46, | |
| "learning_rate": 9.777503090234858e-05, | |
| "loss": 0.0321, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 23.46, | |
| "eval_loss": 0.3872096538543701, | |
| "eval_runtime": 181.6396, | |
| "eval_samples_per_second": 17.904, | |
| "eval_steps_per_second": 2.241, | |
| "eval_wer": 0.23847415416388265, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 24.69, | |
| "learning_rate": 9.765142150803462e-05, | |
| "loss": 0.0342, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 24.69, | |
| "eval_loss": 0.4173298180103302, | |
| "eval_runtime": 185.9384, | |
| "eval_samples_per_second": 17.49, | |
| "eval_steps_per_second": 2.189, | |
| "eval_wer": 0.24266537083156167, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 25.93, | |
| "learning_rate": 9.752781211372065e-05, | |
| "loss": 0.0348, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 25.93, | |
| "eval_loss": 0.36838769912719727, | |
| "eval_runtime": 184.1892, | |
| "eval_samples_per_second": 17.656, | |
| "eval_steps_per_second": 2.21, | |
| "eval_wer": 0.23671262831804654, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 27.16, | |
| "learning_rate": 9.740420271940668e-05, | |
| "loss": 0.0332, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 27.16, | |
| "eval_loss": 0.3941015899181366, | |
| "eval_runtime": 184.3624, | |
| "eval_samples_per_second": 17.639, | |
| "eval_steps_per_second": 2.208, | |
| "eval_wer": 0.23233918483872928, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 28.4, | |
| "learning_rate": 9.728059332509271e-05, | |
| "loss": 0.0339, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 28.4, | |
| "eval_loss": 0.3854130506515503, | |
| "eval_runtime": 186.7325, | |
| "eval_samples_per_second": 17.415, | |
| "eval_steps_per_second": 2.18, | |
| "eval_wer": 0.24254388629046955, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 29.63, | |
| "learning_rate": 9.715698393077875e-05, | |
| "loss": 0.0349, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 29.63, | |
| "eval_loss": 0.39330288767814636, | |
| "eval_runtime": 183.7314, | |
| "eval_samples_per_second": 17.7, | |
| "eval_steps_per_second": 2.215, | |
| "eval_wer": 0.23774524691732976, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 30.86, | |
| "learning_rate": 9.703337453646477e-05, | |
| "loss": 0.0327, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 30.86, | |
| "eval_loss": 0.3882978558540344, | |
| "eval_runtime": 186.0182, | |
| "eval_samples_per_second": 17.482, | |
| "eval_steps_per_second": 2.188, | |
| "eval_wer": 0.2347081333900261, | |
| "step": 2500 | |
| } | |
| ], | |
| "max_steps": 81000, | |
| "num_train_epochs": 1000, | |
| "total_flos": 7.954928320372541e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |