| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5952380952380952, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.023809523809523808, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.023809523809523808, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.1589, | |
| "eval_samples_per_second": 31.064, | |
| "eval_steps_per_second": 7.766, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.047619047619047616, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.07142857142857142, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.07142857142857142, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.1881, | |
| "eval_samples_per_second": 30.3, | |
| "eval_steps_per_second": 7.575, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.11904761904761904, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.14285714285714285, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.1268, | |
| "eval_samples_per_second": 31.948, | |
| "eval_steps_per_second": 7.987, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.21428571428571427, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.21428571428571427, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.1252, | |
| "eval_samples_per_second": 31.994, | |
| "eval_steps_per_second": 7.999, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.23809523809523808, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2619047619047619, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.124, | |
| "eval_samples_per_second": 32.029, | |
| "eval_steps_per_second": 8.007, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.30952380952380953, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.35714285714285715, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.35714285714285715, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.127, | |
| "eval_samples_per_second": 31.944, | |
| "eval_steps_per_second": 7.986, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.40476190476190477, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.1357, | |
| "eval_samples_per_second": 31.697, | |
| "eval_steps_per_second": 7.924, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.4523809523809524, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.1239, | |
| "eval_samples_per_second": 32.03, | |
| "eval_steps_per_second": 8.008, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.5238095238095238, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.5476190476190477, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 0.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.1236, | |
| "eval_samples_per_second": 32.04, | |
| "eval_steps_per_second": 8.01, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5952380952380952, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.69912849629184e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |