| { | |
| "best_metric": 21.407068619939793, | |
| "best_model_checkpoint": "/cosmos/home/sp-operator/ai/training/models/huggingface/scripts/../breeze-listen-dsw-base-ta/checkpoint-1000", | |
| "epoch": 1.0, | |
| "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": 0.5836, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 6.195318418690893e-06, | |
| "loss": 0.308, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.881634451095711e-06, | |
| "loss": 0.2177, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 7.361221988663844e-06, | |
| "loss": 0.1698, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.572265625, | |
| "eval_runtime": 14672.9249, | |
| "eval_samples_per_second": 0.79, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 30.440569885233597, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 7.730207550743121e-06, | |
| "loss": 0.1375, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.03016458599496e-06, | |
| "loss": 0.4086, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.282894746203441e-06, | |
| "loss": 0.4454, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.501266121799902e-06, | |
| "loss": 0.3578, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.43017578125, | |
| "eval_runtime": 14830.0095, | |
| "eval_samples_per_second": 0.782, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 25.686238091921464, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.693512601774437e-06, | |
| "loss": 0.3099, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 8.865222471593567e-06, | |
| "loss": 0.3064, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 9.020362953730323e-06, | |
| "loss": 0.2555, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 9.161852281961698e-06, | |
| "loss": 0.2832, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.396728515625, | |
| "eval_runtime": 14915.7095, | |
| "eval_samples_per_second": 0.778, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 23.204791864973416, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 9.29189975311636e-06, | |
| "loss": 0.239, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 9.412218256259678e-06, | |
| "loss": 0.2391, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 9.524162683365145e-06, | |
| "loss": 0.3022, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 9.62882322733502e-06, | |
| "loss": 0.2663, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.40380859375, | |
| "eval_runtime": 14958.2575, | |
| "eval_samples_per_second": 0.775, | |
| "eval_steps_per_second": 0.048, | |
| "eval_wer": 23.852529426787083, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 9.727090137141168e-06, | |
| "loss": 0.2667, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 9.819699807237934e-06, | |
| "loss": 0.2368, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 9.907268307310855e-06, | |
| "loss": 0.4503, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 9.990316248055788e-06, | |
| "loss": 0.5175, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.396240234375, | |
| "eval_runtime": 14832.7471, | |
| "eval_samples_per_second": 0.782, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 24.14660382761049, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 9.58e-06, | |
| "loss": 0.4672, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 9.080000000000001e-06, | |
| "loss": 0.4226, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 8.580000000000001e-06, | |
| "loss": 0.3758, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.08e-06, | |
| "loss": 0.2365, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.385009765625, | |
| "eval_runtime": 14663.0923, | |
| "eval_samples_per_second": 0.791, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 22.25949744232659, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 7.58e-06, | |
| "loss": 0.205, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 7.08e-06, | |
| "loss": 0.2127, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 6.5800000000000005e-06, | |
| "loss": 0.1918, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 6.08e-06, | |
| "loss": 0.1692, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.39599609375, | |
| "eval_runtime": 14696.2041, | |
| "eval_samples_per_second": 0.789, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 21.868688041232325, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 5.580000000000001e-06, | |
| "loss": 0.155, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 5.0800000000000005e-06, | |
| "loss": 0.1639, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.58e-06, | |
| "loss": 0.1611, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.08e-06, | |
| "loss": 0.1815, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.38232421875, | |
| "eval_runtime": 14658.9058, | |
| "eval_samples_per_second": 0.791, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 22.077248701816295, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 3.58e-06, | |
| "loss": 0.1854, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.08e-06, | |
| "loss": 0.175, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 2.5800000000000003e-06, | |
| "loss": 0.1684, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.08e-06, | |
| "loss": 0.1612, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.3701171875, | |
| "eval_runtime": 14864.357, | |
| "eval_samples_per_second": 0.78, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 21.805616821055725, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1.5800000000000001e-06, | |
| "loss": 0.1528, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.08e-06, | |
| "loss": 0.1651, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 5.800000000000001e-07, | |
| "loss": 0.1474, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 8e-08, | |
| "loss": 0.1393, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.375, | |
| "eval_runtime": 14729.6157, | |
| "eval_samples_per_second": 0.787, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 21.407068619939793, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 1000, | |
| "total_flos": 2.0755199247672934e+18, | |
| "train_loss": 0.2634566650390625, | |
| "train_runtime": 177344.9657, | |
| "train_samples_per_second": 0.18, | |
| "train_steps_per_second": 0.006 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 100, | |
| "total_flos": 2.0755199247672934e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |