| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 31.71508379888268, | |
| "global_step": 4250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0641, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 4.999626837823718e-05, | |
| "loss": 0.041, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_loss": 0.1974974274635315, | |
| "eval_runtime": 282.4584, | |
| "eval_samples_per_second": 13.57, | |
| "eval_steps_per_second": 1.699, | |
| "eval_wer": 0.18217074305806918, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 4.999253675647437e-05, | |
| "loss": 0.0381, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 4.998884245092918e-05, | |
| "loss": 0.0343, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 4.998511082916636e-05, | |
| "loss": 0.0325, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "eval_loss": 0.2165367156267166, | |
| "eval_runtime": 199.6867, | |
| "eval_samples_per_second": 19.195, | |
| "eval_steps_per_second": 2.404, | |
| "eval_wer": 0.18279125084027095, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 4.998137920740354e-05, | |
| "loss": 0.0324, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 4.9977647585640724e-05, | |
| "loss": 0.0303, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "eval_loss": 0.23278702795505524, | |
| "eval_runtime": 211.6975, | |
| "eval_samples_per_second": 18.106, | |
| "eval_steps_per_second": 2.267, | |
| "eval_wer": 0.1827395418584208, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 4.9973915963877904e-05, | |
| "loss": 0.031, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 4.9970184342115084e-05, | |
| "loss": 0.0287, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 4.996645272035227e-05, | |
| "loss": 0.0281, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "eval_loss": 0.23425738513469696, | |
| "eval_runtime": 209.1881, | |
| "eval_samples_per_second": 18.323, | |
| "eval_steps_per_second": 2.295, | |
| "eval_wer": 0.18170536222141787, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 4.996272109858945e-05, | |
| "loss": 0.0266, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 4.995898947682663e-05, | |
| "loss": 0.0263, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "eval_loss": 0.23851124942302704, | |
| "eval_runtime": 208.1334, | |
| "eval_samples_per_second": 18.416, | |
| "eval_steps_per_second": 2.306, | |
| "eval_wer": 0.17855111432855886, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 9.7, | |
| "learning_rate": 4.995525785506381e-05, | |
| "loss": 0.0265, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.45, | |
| "learning_rate": 4.995152623330099e-05, | |
| "loss": 0.0263, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "learning_rate": 4.994779461153818e-05, | |
| "loss": 0.0262, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "eval_loss": 0.2488125115633011, | |
| "eval_runtime": 216.4749, | |
| "eval_samples_per_second": 17.706, | |
| "eval_steps_per_second": 2.217, | |
| "eval_wer": 0.17699984487305445, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "learning_rate": 4.994406298977536e-05, | |
| "loss": 0.0237, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 4.994033136801254e-05, | |
| "loss": 0.0249, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "eval_loss": 0.261038601398468, | |
| "eval_runtime": 219.4529, | |
| "eval_samples_per_second": 17.466, | |
| "eval_steps_per_second": 2.187, | |
| "eval_wer": 0.17513832152644915, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 13.43, | |
| "learning_rate": 4.9936599746249724e-05, | |
| "loss": 0.0235, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "learning_rate": 4.9932868124486904e-05, | |
| "loss": 0.0246, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 14.92, | |
| "learning_rate": 4.992913650272409e-05, | |
| "loss": 0.0227, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 14.92, | |
| "eval_loss": 0.25547119975090027, | |
| "eval_runtime": 215.7055, | |
| "eval_samples_per_second": 17.77, | |
| "eval_steps_per_second": 2.225, | |
| "eval_wer": 0.17472464967164797, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.67, | |
| "learning_rate": 4.992540488096127e-05, | |
| "loss": 0.0215, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 16.42, | |
| "learning_rate": 4.992167325919845e-05, | |
| "loss": 0.0218, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 16.79, | |
| "eval_loss": 0.25752121210098267, | |
| "eval_runtime": 207.5079, | |
| "eval_samples_per_second": 18.472, | |
| "eval_steps_per_second": 2.313, | |
| "eval_wer": 0.17425926883499665, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 17.16, | |
| "learning_rate": 4.991794163743563e-05, | |
| "loss": 0.0213, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 17.91, | |
| "learning_rate": 4.991421001567281e-05, | |
| "loss": 0.0212, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 18.66, | |
| "learning_rate": 4.991047839390999e-05, | |
| "loss": 0.0211, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 18.66, | |
| "eval_loss": 0.2587244212627411, | |
| "eval_runtime": 211.1091, | |
| "eval_samples_per_second": 18.156, | |
| "eval_steps_per_second": 2.274, | |
| "eval_wer": 0.1761725011634521, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 19.4, | |
| "learning_rate": 4.990674677214718e-05, | |
| "loss": 0.0201, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 20.15, | |
| "learning_rate": 4.990301515038436e-05, | |
| "loss": 0.0222, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 20.52, | |
| "eval_loss": 0.26894259452819824, | |
| "eval_runtime": 212.0938, | |
| "eval_samples_per_second": 18.072, | |
| "eval_steps_per_second": 2.263, | |
| "eval_wer": 0.17451781374424738, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 20.89, | |
| "learning_rate": 4.9899283528621545e-05, | |
| "loss": 0.0207, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 21.64, | |
| "learning_rate": 4.9895551906858725e-05, | |
| "loss": 0.0194, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 22.39, | |
| "learning_rate": 4.9891820285095905e-05, | |
| "loss": 0.021, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 22.39, | |
| "eval_loss": 0.26362183690071106, | |
| "eval_runtime": 204.8903, | |
| "eval_samples_per_second": 18.708, | |
| "eval_steps_per_second": 2.343, | |
| "eval_wer": 0.1751900305082993, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 23.13, | |
| "learning_rate": 4.9888088663333085e-05, | |
| "loss": 0.0216, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 23.88, | |
| "learning_rate": 4.9884357041570265e-05, | |
| "loss": 0.02, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 24.25, | |
| "eval_loss": 0.2709738612174988, | |
| "eval_runtime": 211.3483, | |
| "eval_samples_per_second": 18.136, | |
| "eval_steps_per_second": 2.271, | |
| "eval_wer": 0.17513832152644915, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 24.63, | |
| "learning_rate": 4.988062541980745e-05, | |
| "loss": 0.02, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 25.37, | |
| "learning_rate": 4.987689379804463e-05, | |
| "loss": 0.0206, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 26.12, | |
| "learning_rate": 4.987316217628181e-05, | |
| "loss": 0.0195, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 26.12, | |
| "eval_loss": 0.26540952920913696, | |
| "eval_runtime": 209.7499, | |
| "eval_samples_per_second": 18.274, | |
| "eval_steps_per_second": 2.288, | |
| "eval_wer": 0.17570712032680078, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 26.86, | |
| "learning_rate": 4.9869430554519e-05, | |
| "loss": 0.0193, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 27.61, | |
| "learning_rate": 4.986569893275618e-05, | |
| "loss": 0.0197, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 27.98, | |
| "eval_loss": 0.26861485838890076, | |
| "eval_runtime": 224.5, | |
| "eval_samples_per_second": 17.073, | |
| "eval_steps_per_second": 2.138, | |
| "eval_wer": 0.17482806763534825, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 28.36, | |
| "learning_rate": 4.9861967310993365e-05, | |
| "loss": 0.0198, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 29.1, | |
| "learning_rate": 4.9858235689230545e-05, | |
| "loss": 0.0199, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 29.85, | |
| "learning_rate": 4.9854504067467725e-05, | |
| "loss": 0.0187, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 29.85, | |
| "eval_loss": 0.2687038481235504, | |
| "eval_runtime": 221.4533, | |
| "eval_samples_per_second": 17.308, | |
| "eval_steps_per_second": 2.168, | |
| "eval_wer": 0.1751900305082993, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 30.6, | |
| "learning_rate": 4.9850772445704905e-05, | |
| "loss": 0.019, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 31.34, | |
| "learning_rate": 4.9847040823942085e-05, | |
| "loss": 0.0188, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 31.72, | |
| "eval_loss": 0.2625608444213867, | |
| "eval_runtime": 222.8747, | |
| "eval_samples_per_second": 17.198, | |
| "eval_steps_per_second": 2.154, | |
| "eval_wer": 0.1721392005791406, | |
| "step": 4250 | |
| } | |
| ], | |
| "max_steps": 1340000, | |
| "num_train_epochs": 10000, | |
| "total_flos": 9.643398954166528e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |