| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 37.001, |
| "eval_steps": 1000, |
| "global_step": 1000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.025, |
| "grad_norm": 13.28831958770752, |
| "learning_rate": 4.4e-07, |
| "loss": 1.7404, |
| "step": 25 |
| }, |
| { |
| "epoch": 1.023, |
| "grad_norm": 9.838354110717773, |
| "learning_rate": 9.400000000000001e-07, |
| "loss": 1.5694, |
| "step": 50 |
| }, |
| { |
| "epoch": 2.021, |
| "grad_norm": 8.474053382873535, |
| "learning_rate": 1.44e-06, |
| "loss": 1.3389, |
| "step": 75 |
| }, |
| { |
| "epoch": 3.019, |
| "grad_norm": 10.3988037109375, |
| "learning_rate": 1.94e-06, |
| "loss": 1.1896, |
| "step": 100 |
| }, |
| { |
| "epoch": 4.017, |
| "grad_norm": 10.84144401550293, |
| "learning_rate": 2.4400000000000004e-06, |
| "loss": 1.0255, |
| "step": 125 |
| }, |
| { |
| "epoch": 5.015, |
| "grad_norm": 9.548709869384766, |
| "learning_rate": 2.9400000000000002e-06, |
| "loss": 0.8268, |
| "step": 150 |
| }, |
| { |
| "epoch": 6.013, |
| "grad_norm": 10.803505897521973, |
| "learning_rate": 3.44e-06, |
| "loss": 0.6162, |
| "step": 175 |
| }, |
| { |
| "epoch": 7.011, |
| "grad_norm": 11.32190990447998, |
| "learning_rate": 3.94e-06, |
| "loss": 0.4587, |
| "step": 200 |
| }, |
| { |
| "epoch": 8.009, |
| "grad_norm": 9.957799911499023, |
| "learning_rate": 4.440000000000001e-06, |
| "loss": 0.3452, |
| "step": 225 |
| }, |
| { |
| "epoch": 9.007, |
| "grad_norm": 8.934591293334961, |
| "learning_rate": 4.94e-06, |
| "loss": 0.2472, |
| "step": 250 |
| }, |
| { |
| "epoch": 10.005, |
| "grad_norm": 5.261079788208008, |
| "learning_rate": 5.4400000000000004e-06, |
| "loss": 0.1786, |
| "step": 275 |
| }, |
| { |
| "epoch": 11.003, |
| "grad_norm": 5.313260555267334, |
| "learning_rate": 5.94e-06, |
| "loss": 0.1278, |
| "step": 300 |
| }, |
| { |
| "epoch": 12.001, |
| "grad_norm": 5.021953582763672, |
| "learning_rate": 6.440000000000001e-06, |
| "loss": 0.0851, |
| "step": 325 |
| }, |
| { |
| "epoch": 12.026, |
| "grad_norm": 6.381885051727295, |
| "learning_rate": 6.9400000000000005e-06, |
| "loss": 0.0628, |
| "step": 350 |
| }, |
| { |
| "epoch": 13.024, |
| "grad_norm": 8.314247131347656, |
| "learning_rate": 7.440000000000001e-06, |
| "loss": 0.0428, |
| "step": 375 |
| }, |
| { |
| "epoch": 14.022, |
| "grad_norm": 7.420034885406494, |
| "learning_rate": 7.94e-06, |
| "loss": 0.0384, |
| "step": 400 |
| }, |
| { |
| "epoch": 15.02, |
| "grad_norm": 4.205871105194092, |
| "learning_rate": 8.44e-06, |
| "loss": 0.0304, |
| "step": 425 |
| }, |
| { |
| "epoch": 16.018, |
| "grad_norm": 5.158091068267822, |
| "learning_rate": 8.94e-06, |
| "loss": 0.0307, |
| "step": 450 |
| }, |
| { |
| "epoch": 17.016, |
| "grad_norm": 3.9861438274383545, |
| "learning_rate": 9.440000000000001e-06, |
| "loss": 0.0255, |
| "step": 475 |
| }, |
| { |
| "epoch": 18.014, |
| "grad_norm": 6.794647693634033, |
| "learning_rate": 9.940000000000001e-06, |
| "loss": 0.0297, |
| "step": 500 |
| }, |
| { |
| "epoch": 19.012, |
| "grad_norm": 2.1698760986328125, |
| "learning_rate": 9.56e-06, |
| "loss": 0.0209, |
| "step": 525 |
| }, |
| { |
| "epoch": 20.01, |
| "grad_norm": 2.9011716842651367, |
| "learning_rate": 9.060000000000001e-06, |
| "loss": 0.0226, |
| "step": 550 |
| }, |
| { |
| "epoch": 21.008, |
| "grad_norm": 4.619945526123047, |
| "learning_rate": 8.560000000000001e-06, |
| "loss": 0.0203, |
| "step": 575 |
| }, |
| { |
| "epoch": 22.006, |
| "grad_norm": 1.794995903968811, |
| "learning_rate": 8.06e-06, |
| "loss": 0.0114, |
| "step": 600 |
| }, |
| { |
| "epoch": 23.004, |
| "grad_norm": 2.5991861820220947, |
| "learning_rate": 7.5600000000000005e-06, |
| "loss": 0.0115, |
| "step": 625 |
| }, |
| { |
| "epoch": 24.002, |
| "grad_norm": 3.6725001335144043, |
| "learning_rate": 7.06e-06, |
| "loss": 0.0112, |
| "step": 650 |
| }, |
| { |
| "epoch": 24.027, |
| "grad_norm": 3.874669313430786, |
| "learning_rate": 6.560000000000001e-06, |
| "loss": 0.0072, |
| "step": 675 |
| }, |
| { |
| "epoch": 25.025, |
| "grad_norm": 4.749724864959717, |
| "learning_rate": 6.0600000000000004e-06, |
| "loss": 0.0052, |
| "step": 700 |
| }, |
| { |
| "epoch": 26.023, |
| "grad_norm": 2.345092296600342, |
| "learning_rate": 5.560000000000001e-06, |
| "loss": 0.0044, |
| "step": 725 |
| }, |
| { |
| "epoch": 27.021, |
| "grad_norm": 2.500375986099243, |
| "learning_rate": 5.060000000000001e-06, |
| "loss": 0.0041, |
| "step": 750 |
| }, |
| { |
| "epoch": 28.019, |
| "grad_norm": 2.1209278106689453, |
| "learning_rate": 4.56e-06, |
| "loss": 0.0027, |
| "step": 775 |
| }, |
| { |
| "epoch": 29.017, |
| "grad_norm": 0.10120876878499985, |
| "learning_rate": 4.060000000000001e-06, |
| "loss": 0.002, |
| "step": 800 |
| }, |
| { |
| "epoch": 30.015, |
| "grad_norm": 0.04560551419854164, |
| "learning_rate": 3.5600000000000002e-06, |
| "loss": 0.0008, |
| "step": 825 |
| }, |
| { |
| "epoch": 31.013, |
| "grad_norm": 0.042017169296741486, |
| "learning_rate": 3.0600000000000003e-06, |
| "loss": 0.0004, |
| "step": 850 |
| }, |
| { |
| "epoch": 32.011, |
| "grad_norm": 0.0233773086220026, |
| "learning_rate": 2.56e-06, |
| "loss": 0.0005, |
| "step": 875 |
| }, |
| { |
| "epoch": 33.009, |
| "grad_norm": 0.03428841382265091, |
| "learning_rate": 2.06e-06, |
| "loss": 0.0004, |
| "step": 900 |
| }, |
| { |
| "epoch": 34.007, |
| "grad_norm": 0.018520979210734367, |
| "learning_rate": 1.56e-06, |
| "loss": 0.0003, |
| "step": 925 |
| }, |
| { |
| "epoch": 35.005, |
| "grad_norm": 0.016791554167866707, |
| "learning_rate": 1.06e-06, |
| "loss": 0.0002, |
| "step": 950 |
| }, |
| { |
| "epoch": 36.003, |
| "grad_norm": 0.0166826993227005, |
| "learning_rate": 5.6e-07, |
| "loss": 0.0002, |
| "step": 975 |
| }, |
| { |
| "epoch": 37.001, |
| "grad_norm": 0.015415076166391373, |
| "learning_rate": 6.000000000000001e-08, |
| "loss": 0.0002, |
| "step": 1000 |
| }, |
| { |
| "epoch": 37.001, |
| "eval_loss": 0.5906396508216858, |
| "eval_runtime": 51.9408, |
| "eval_samples_per_second": 2.156, |
| "eval_steps_per_second": 0.27, |
| "eval_wer": 38.88888888888889, |
| "step": 1000 |
| }, |
| { |
| "epoch": 37.001, |
| "step": 1000, |
| "total_flos": 4.59601108549632e+18, |
| "train_loss": 0.2534070857851766, |
| "train_runtime": 2799.3838, |
| "train_samples_per_second": 5.716, |
| "train_steps_per_second": 0.357 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 1000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9223372036854775807, |
| "save_steps": 1000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4.59601108549632e+18, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
