| { | |
| "best_metric": 43.44465912227436, | |
| "best_model_checkpoint": "/cosmos/home/sp-operator/ai/training/models/huggingface/scripts/../breeze-dsw-tiny-id/checkpoint-900", | |
| "epoch": 6.064, | |
| "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": 4.898977360288234e-06, | |
| "loss": 1.7848, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 6.160712527409633e-06, | |
| "loss": 1.3566, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.85912902234906e-06, | |
| "loss": 1.1406, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 7.344547104469332e-06, | |
| "loss": 0.99, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.8486328125, | |
| "eval_runtime": 1287.4485, | |
| "eval_samples_per_second": 2.829, | |
| "eval_steps_per_second": 0.177, | |
| "eval_wer": 54.24602079308124, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 7.716963756434345e-06, | |
| "loss": 0.765, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.019180844200955e-06, | |
| "loss": 0.5849, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 8.27351214279797e-06, | |
| "loss": 0.7834, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 8.49307723936858e-06, | |
| "loss": 0.7896, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "eval_loss": 0.7578125, | |
| "eval_runtime": 1159.7848, | |
| "eval_samples_per_second": 3.14, | |
| "eval_steps_per_second": 0.197, | |
| "eval_wer": 48.389916275646335, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 8.686247975778677e-06, | |
| "loss": 0.8414, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 8.858694625217149e-06, | |
| "loss": 0.6799, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 9.014436199608479e-06, | |
| "loss": 0.5851, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 9.156425255148058e-06, | |
| "loss": 0.4164, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_loss": 0.73876953125, | |
| "eval_runtime": 1190.5006, | |
| "eval_samples_per_second": 3.059, | |
| "eval_steps_per_second": 0.192, | |
| "eval_wer": 49.3283650749839, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.28689473531776e-06, | |
| "loss": 0.5103, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.407574351377137e-06, | |
| "loss": 0.6019, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 9.519831289296397e-06, | |
| "loss": 0.6399, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.624764935335318e-06, | |
| "loss": 0.5456, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "eval_loss": 0.7177734375, | |
| "eval_runtime": 1156.6175, | |
| "eval_samples_per_second": 3.149, | |
| "eval_steps_per_second": 0.197, | |
| "eval_wer": 45.795381359830714, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 9.723272550712454e-06, | |
| "loss": 0.48, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 9.816095971633122e-06, | |
| "loss": 0.3325, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 9.90385555539545e-06, | |
| "loss": 0.3389, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 9.987075336738768e-06, | |
| "loss": 0.4761, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "eval_loss": 0.7109375, | |
| "eval_runtime": 1128.5074, | |
| "eval_samples_per_second": 3.227, | |
| "eval_steps_per_second": 0.202, | |
| "eval_wer": 45.21575121906339, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.4891, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 9.100000000000001e-06, | |
| "loss": 0.4411, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 8.6e-06, | |
| "loss": 0.367, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 8.1e-06, | |
| "loss": 0.2674, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "eval_loss": 0.70068359375, | |
| "eval_runtime": 1151.1688, | |
| "eval_samples_per_second": 3.164, | |
| "eval_steps_per_second": 0.198, | |
| "eval_wer": 44.843131842855826, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 0.2263, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 7.100000000000001e-06, | |
| "loss": 0.3527, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 6.600000000000001e-06, | |
| "loss": 0.369, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 6.1e-06, | |
| "loss": 0.3628, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "eval_loss": 0.70263671875, | |
| "eval_runtime": 1168.9802, | |
| "eval_samples_per_second": 3.116, | |
| "eval_steps_per_second": 0.195, | |
| "eval_wer": 44.24970098445119, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 0.2981, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 5.1e-06, | |
| "loss": 0.2296, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 4.600000000000001e-06, | |
| "loss": 0.1694, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 4.1e-06, | |
| "loss": 0.2565, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "eval_loss": 0.70849609375, | |
| "eval_runtime": 1151.0035, | |
| "eval_samples_per_second": 3.164, | |
| "eval_steps_per_second": 0.198, | |
| "eval_wer": 44.50731438034778, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "loss": 0.2965, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 3.1000000000000004e-06, | |
| "loss": 0.2987, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 2.6e-06, | |
| "loss": 0.2593, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 2.1000000000000002e-06, | |
| "loss": 0.2147, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "eval_loss": 0.708984375, | |
| "eval_runtime": 1111.0874, | |
| "eval_samples_per_second": 3.278, | |
| "eval_steps_per_second": 0.205, | |
| "eval_wer": 43.44465912227436, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 0.1495, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 1.1e-06, | |
| "loss": 0.1969, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 0.2518, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 1.0000000000000001e-07, | |
| "loss": 0.28, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "eval_loss": 0.71337890625, | |
| "eval_runtime": 1142.0113, | |
| "eval_samples_per_second": 3.189, | |
| "eval_steps_per_second": 0.2, | |
| "eval_wer": 43.95528567485509, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "step": 1000, | |
| "total_flos": 7.844068896339395e+17, | |
| "train_loss": 0.5104879150390625, | |
| "train_runtime": 24718.0782, | |
| "train_samples_per_second": 1.295, | |
| "train_steps_per_second": 0.04 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 100, | |
| "total_flos": 7.844068896339395e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |