| { | |
| "best_metric": 6.624661667480143, | |
| "best_model_checkpoint": "whisper-stakwork-crypto/checkpoint-275", | |
| "epoch": 16.04, | |
| "global_step": 350, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 8e-09, | |
| "loss": 0.8207, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.1333333333333332e-08, | |
| "loss": 0.8163, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 0.8355672955513, | |
| "eval_runtime": 119.9965, | |
| "eval_samples_per_second": 2.283, | |
| "eval_steps_per_second": 0.15, | |
| "eval_wer": 10.063451213559924, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 3.4666666666666666e-08, | |
| "loss": 0.8041, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 4.799999999999999e-08, | |
| "loss": 0.8138, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 6.133333333333333e-08, | |
| "loss": 0.7916, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_loss": 0.8102102875709534, | |
| "eval_runtime": 128.996, | |
| "eval_samples_per_second": 2.124, | |
| "eval_steps_per_second": 0.14, | |
| "eval_wer": 9.735102276256823, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 7.466666666666666e-08, | |
| "loss": 0.7937, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 8.8e-08, | |
| "loss": 0.7563, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "eval_loss": 0.7621258497238159, | |
| "eval_runtime": 118.1768, | |
| "eval_samples_per_second": 2.319, | |
| "eval_steps_per_second": 0.152, | |
| "eval_wer": 9.508807738385766, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 1.0133333333333334e-07, | |
| "loss": 0.7477, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 1.1466666666666666e-07, | |
| "loss": 0.7275, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 1.28e-07, | |
| "loss": 0.7154, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "eval_loss": 0.7107397317886353, | |
| "eval_runtime": 132.2935, | |
| "eval_samples_per_second": 2.071, | |
| "eval_steps_per_second": 0.136, | |
| "eval_wer": 9.233704574699383, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 1.4133333333333333e-07, | |
| "loss": 0.6774, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 1.5466666666666665e-07, | |
| "loss": 0.6548, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "eval_loss": 0.6588723659515381, | |
| "eval_runtime": 118.5646, | |
| "eval_samples_per_second": 2.311, | |
| "eval_steps_per_second": 0.152, | |
| "eval_wer": 9.38013045214536, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 1.68e-07, | |
| "loss": 0.6362, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 1.8133333333333332e-07, | |
| "loss": 0.6105, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 1.9466666666666666e-07, | |
| "loss": 0.6017, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "eval_loss": 0.6061531901359558, | |
| "eval_runtime": 125.9398, | |
| "eval_samples_per_second": 2.176, | |
| "eval_steps_per_second": 0.143, | |
| "eval_wer": 9.007410036828327, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 2.08e-07, | |
| "loss": 0.5569, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 2.2133333333333333e-07, | |
| "loss": 0.5333, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "eval_loss": 0.5347172617912292, | |
| "eval_runtime": 129.1421, | |
| "eval_samples_per_second": 2.122, | |
| "eval_steps_per_second": 0.139, | |
| "eval_wer": 8.621378178107113, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 2.3466666666666665e-07, | |
| "loss": 0.4959, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2.48e-07, | |
| "loss": 0.4715, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 2.613333333333333e-07, | |
| "loss": 0.4493, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "eval_loss": 0.47380271553993225, | |
| "eval_runtime": 131.2715, | |
| "eval_samples_per_second": 2.087, | |
| "eval_steps_per_second": 0.137, | |
| "eval_wer": 8.284154945201225, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 2.7466666666666664e-07, | |
| "loss": 0.4248, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 2.88e-07, | |
| "loss": 0.4016, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "eval_loss": 0.43325528502464294, | |
| "eval_runtime": 126.954, | |
| "eval_samples_per_second": 2.158, | |
| "eval_steps_per_second": 0.142, | |
| "eval_wer": 7.117185073434797, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 3.0133333333333333e-07, | |
| "loss": 0.4003, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 3.146666666666666e-07, | |
| "loss": 0.374, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 3.2799999999999997e-07, | |
| "loss": 0.3738, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "eval_loss": 0.40573614835739136, | |
| "eval_runtime": 120.1397, | |
| "eval_samples_per_second": 2.281, | |
| "eval_steps_per_second": 0.15, | |
| "eval_wer": 6.7000931801038295, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 3.413333333333333e-07, | |
| "loss": 0.346, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 3.5466666666666667e-07, | |
| "loss": 0.3544, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "eval_loss": 0.38816705346107483, | |
| "eval_runtime": 121.943, | |
| "eval_samples_per_second": 2.247, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 6.624661667480143, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 3.68e-07, | |
| "loss": 0.3345, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 3.813333333333333e-07, | |
| "loss": 0.3354, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 3.9466666666666665e-07, | |
| "loss": 0.3294, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "eval_loss": 0.37637442350387573, | |
| "eval_runtime": 132.1904, | |
| "eval_samples_per_second": 2.073, | |
| "eval_steps_per_second": 0.136, | |
| "eval_wer": 6.695656032302436, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "learning_rate": 3.52e-07, | |
| "loss": 0.321, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 2.72e-07, | |
| "loss": 0.313, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "eval_loss": 0.3692246973514557, | |
| "eval_runtime": 129.104, | |
| "eval_samples_per_second": 2.122, | |
| "eval_steps_per_second": 0.139, | |
| "eval_wer": 6.66015884989129, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 15.04, | |
| "learning_rate": 1.9199999999999997e-07, | |
| "loss": 0.3141, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 1.1200000000000001e-07, | |
| "loss": 0.3172, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 3.2e-08, | |
| "loss": 0.3023, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "eval_loss": 0.36682912707328796, | |
| "eval_runtime": 127.3924, | |
| "eval_samples_per_second": 2.151, | |
| "eval_steps_per_second": 0.141, | |
| "eval_wer": 6.646847406487109, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "step": 350, | |
| "total_flos": 1.125114850050048e+19, | |
| "train_loss": 0.5347552497046334, | |
| "train_runtime": 7389.0047, | |
| "train_samples_per_second": 1.516, | |
| "train_steps_per_second": 0.047 | |
| } | |
| ], | |
| "max_steps": 350, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.125114850050048e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |