| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.21164021164021163, | |
| "eval_steps": 1000, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.25e-08, | |
| "loss": 0.6602, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 2.5e-08, | |
| "loss": 1.0487, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 3.75e-08, | |
| "loss": 0.6436, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 5e-08, | |
| "loss": 1.0184, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 6.25e-08, | |
| "loss": 0.6427, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 7.5e-08, | |
| "loss": 0.9576, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 8.75e-08, | |
| "loss": 0.5878, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1e-07, | |
| "loss": 0.9261, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.125e-07, | |
| "loss": 0.5172, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.25e-07, | |
| "loss": 0.7859, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.375e-07, | |
| "loss": 0.4913, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.5e-07, | |
| "loss": 0.6184, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.625e-07, | |
| "loss": 0.3088, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.75e-07, | |
| "loss": 0.3973, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.875e-07, | |
| "loss": 0.2386, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 2e-07, | |
| "loss": 0.2516, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 2.1249999999999998e-07, | |
| "loss": 0.168, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.25e-07, | |
| "loss": 0.2036, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.3749999999999998e-07, | |
| "loss": 0.1564, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2.5e-07, | |
| "loss": 0.2491, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2.625e-07, | |
| "loss": 0.1817, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.75e-07, | |
| "loss": 0.236, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.8749999999999995e-07, | |
| "loss": 0.1117, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 3e-07, | |
| "loss": 0.1665, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 3.1249999999999997e-07, | |
| "loss": 0.1533, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3.25e-07, | |
| "loss": 0.2006, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3.375e-07, | |
| "loss": 0.1425, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3.5e-07, | |
| "loss": 0.1769, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3.6249999999999997e-07, | |
| "loss": 0.1259, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.75e-07, | |
| "loss": 0.2038, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.875e-07, | |
| "loss": 0.1431, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4e-07, | |
| "loss": 0.1459, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.1249999999999997e-07, | |
| "loss": 0.1219, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.2499999999999995e-07, | |
| "loss": 0.2224, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.375e-07, | |
| "loss": 0.1409, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.5e-07, | |
| "loss": 0.1993, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.625e-07, | |
| "loss": 0.1486, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.7499999999999995e-07, | |
| "loss": 0.1924, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.875e-07, | |
| "loss": 0.107, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 5e-07, | |
| "loss": 0.1534, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 0.1283227652311325, | |
| "eval_runtime": 271898.2411, | |
| "eval_samples_per_second": 0.035, | |
| "eval_steps_per_second": 0.004, | |
| "eval_wer": 0.08386096256684492, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 1000, | |
| "total_flos": 2.717998645248e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |