| { | |
| "best_metric": 28.50294181738941, | |
| "best_model_checkpoint": "/cosmos/home/sp-operator/ai/training/models/huggingface/scripts/../breeze-listen-dsw-base-hi/checkpoint-800", | |
| "epoch": 8.08, | |
| "eval_steps": 100, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.973833272194737e-06, | |
| "loss": 1.392, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 6.195318418690893e-06, | |
| "loss": 0.8763, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.881634451095711e-06, | |
| "loss": 0.6506, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 7.361221988663844e-06, | |
| "loss": 0.553, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.64453125, | |
| "eval_runtime": 3247.321, | |
| "eval_samples_per_second": 0.867, | |
| "eval_steps_per_second": 0.054, | |
| "eval_wer": 39.498801481804314, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 7.730207550743121e-06, | |
| "loss": 0.5096, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 8.03016458599496e-06, | |
| "loss": 0.4697, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 8.282894746203441e-06, | |
| "loss": 0.4251, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 8.501266121799902e-06, | |
| "loss": 0.3683, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "eval_loss": 0.5341796875, | |
| "eval_runtime": 3136.1352, | |
| "eval_samples_per_second": 0.898, | |
| "eval_steps_per_second": 0.056, | |
| "eval_wer": 33.0660274569623, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 8.693512601774437e-06, | |
| "loss": 0.3335, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 8.865222471593567e-06, | |
| "loss": 0.348, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.020362953730323e-06, | |
| "loss": 0.3151, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 9.161852281961698e-06, | |
| "loss": 0.2855, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_loss": 0.498291015625, | |
| "eval_runtime": 3150.996, | |
| "eval_samples_per_second": 0.894, | |
| "eval_steps_per_second": 0.056, | |
| "eval_wer": 31.42514709086947, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 9.29189975311636e-06, | |
| "loss": 0.2602, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.412218256259678e-06, | |
| "loss": 0.2482, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 9.524162683365145e-06, | |
| "loss": 0.2578, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 9.62882322733502e-06, | |
| "loss": 0.2233, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "eval_loss": 0.48681640625, | |
| "eval_runtime": 3166.0648, | |
| "eval_samples_per_second": 0.889, | |
| "eval_steps_per_second": 0.056, | |
| "eval_wer": 30.154717803443017, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 9.727090137141168e-06, | |
| "loss": 0.2072, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 9.819699807237934e-06, | |
| "loss": 0.1884, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 9.907268307310855e-06, | |
| "loss": 0.1854, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 9.990316248055788e-06, | |
| "loss": 0.1832, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "eval_loss": 0.478271484375, | |
| "eval_runtime": 3160.7276, | |
| "eval_samples_per_second": 0.891, | |
| "eval_steps_per_second": 0.056, | |
| "eval_wer": 28.954020483765525, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 9.58e-06, | |
| "loss": 0.1714, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 9.080000000000001e-06, | |
| "loss": 0.1424, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 8.580000000000001e-06, | |
| "loss": 0.133, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 8.08e-06, | |
| "loss": 0.1431, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "eval_loss": 0.490234375, | |
| "eval_runtime": 3169.5843, | |
| "eval_samples_per_second": 0.888, | |
| "eval_steps_per_second": 0.056, | |
| "eval_wer": 29.182828502941817, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 7.58e-06, | |
| "loss": 0.1226, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 7.08e-06, | |
| "loss": 0.1122, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 6.5800000000000005e-06, | |
| "loss": 0.1021, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 6.08e-06, | |
| "loss": 0.0972, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "eval_loss": 0.5048828125, | |
| "eval_runtime": 3143.5242, | |
| "eval_samples_per_second": 0.896, | |
| "eval_steps_per_second": 0.056, | |
| "eval_wer": 28.63804750490303, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 5.580000000000001e-06, | |
| "loss": 0.0955, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 5.0800000000000005e-06, | |
| "loss": 0.0906, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 4.58e-06, | |
| "loss": 0.0807, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 4.08e-06, | |
| "loss": 0.0715, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "eval_loss": 0.5205078125, | |
| "eval_runtime": 3154.678, | |
| "eval_samples_per_second": 0.893, | |
| "eval_steps_per_second": 0.056, | |
| "eval_wer": 28.50294181738941, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 3.58e-06, | |
| "loss": 0.0744, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 3.08e-06, | |
| "loss": 0.0692, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 2.5800000000000003e-06, | |
| "loss": 0.0661, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 2.08e-06, | |
| "loss": 0.0579, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "eval_loss": 0.53662109375, | |
| "eval_runtime": 3161.7785, | |
| "eval_samples_per_second": 0.891, | |
| "eval_steps_per_second": 0.056, | |
| "eval_wer": 28.94748311178906, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 1.5800000000000001e-06, | |
| "loss": 0.0563, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 1.08e-06, | |
| "loss": 0.0588, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 5.800000000000001e-07, | |
| "loss": 0.0541, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "learning_rate": 8e-08, | |
| "loss": 0.0519, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "eval_loss": 0.5380859375, | |
| "eval_runtime": 3154.4366, | |
| "eval_samples_per_second": 0.893, | |
| "eval_steps_per_second": 0.056, | |
| "eval_wer": 28.7949444323382, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "step": 1000, | |
| "total_flos": 2.0641045545295544e+18, | |
| "train_loss": 0.2532890625, | |
| "train_runtime": 58625.9301, | |
| "train_samples_per_second": 0.546, | |
| "train_steps_per_second": 0.017 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 100, | |
| "total_flos": 2.0641045545295544e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |