| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.119524359703064, | |
| "learning_rate": 0.000196, | |
| "loss": 4.6857, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.163640022277832, | |
| "learning_rate": 0.000192, | |
| "loss": 4.5998, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.1471703052520752, | |
| "learning_rate": 0.000188, | |
| "loss": 4.4927, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.251117467880249, | |
| "learning_rate": 0.00018400000000000003, | |
| "loss": 4.3495, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.2612701654434204, | |
| "learning_rate": 0.00018, | |
| "loss": 4.1938, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.2168090343475342, | |
| "learning_rate": 0.00017600000000000002, | |
| "loss": 4.0735, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.4083530902862549, | |
| "learning_rate": 0.000172, | |
| "loss": 3.9902, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.5325428247451782, | |
| "learning_rate": 0.000168, | |
| "loss": 3.8589, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.7195887565612793, | |
| "learning_rate": 0.000164, | |
| "loss": 3.7472, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.7714509963989258, | |
| "learning_rate": 0.00016, | |
| "loss": 3.7221, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.4559599161148071, | |
| "learning_rate": 0.00015600000000000002, | |
| "loss": 3.5918, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.1422759294509888, | |
| "learning_rate": 0.000152, | |
| "loss": 3.4831, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.813960075378418, | |
| "learning_rate": 0.000148, | |
| "loss": 3.4793, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.7030584216117859, | |
| "learning_rate": 0.000144, | |
| "loss": 3.402, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.7199345827102661, | |
| "learning_rate": 0.00014, | |
| "loss": 3.4112, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.6919622421264648, | |
| "learning_rate": 0.00013600000000000003, | |
| "loss": 3.4019, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.778389036655426, | |
| "learning_rate": 0.000132, | |
| "loss": 3.3161, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.5526257753372192, | |
| "learning_rate": 0.00012800000000000002, | |
| "loss": 3.3214, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.4114692509174347, | |
| "learning_rate": 0.000124, | |
| "loss": 3.3335, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.4152827262878418, | |
| "learning_rate": 0.00012, | |
| "loss": 3.3098, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "grad_norm": 0.4545881748199463, | |
| "learning_rate": 0.000116, | |
| "loss": 3.2991, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 0.4798775017261505, | |
| "learning_rate": 0.00011200000000000001, | |
| "loss": 3.2382, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "grad_norm": 0.49320390820503235, | |
| "learning_rate": 0.00010800000000000001, | |
| "loss": 3.2847, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.4057370722293854, | |
| "learning_rate": 0.00010400000000000001, | |
| "loss": 3.283, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 0.4063659906387329, | |
| "learning_rate": 0.0001, | |
| "loss": 3.2006, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_clap": 0.05581656098365784, | |
| "eval_loss": 4.9824113845825195, | |
| "eval_runtime": 50.0809, | |
| "eval_samples_per_second": 0.04, | |
| "eval_steps_per_second": 0.04, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 0.3333756923675537, | |
| "learning_rate": 9.6e-05, | |
| "loss": 3.2544, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "grad_norm": 0.3300846517086029, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 3.2623, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.3805270195007324, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 3.2162, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "grad_norm": 0.33033835887908936, | |
| "learning_rate": 8.4e-05, | |
| "loss": 3.2294, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.3444644808769226, | |
| "learning_rate": 8e-05, | |
| "loss": 3.2057, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "grad_norm": 0.3191991150379181, | |
| "learning_rate": 7.6e-05, | |
| "loss": 3.2166, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.336597204208374, | |
| "learning_rate": 7.2e-05, | |
| "loss": 3.1764, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "grad_norm": 0.3181450068950653, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 3.2459, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "grad_norm": 0.35749194025993347, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 3.1247, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 0.3413698077201843, | |
| "learning_rate": 6e-05, | |
| "loss": 3.2322, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 0.28314924240112305, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 3.1605, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "grad_norm": 0.3493545353412628, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 3.2135, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "grad_norm": 0.30214983224868774, | |
| "learning_rate": 4.8e-05, | |
| "loss": 3.1751, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "grad_norm": 0.30475127696990967, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 3.1885, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.2956351339817047, | |
| "learning_rate": 4e-05, | |
| "loss": 3.2105, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "grad_norm": 0.2751070559024811, | |
| "learning_rate": 3.6e-05, | |
| "loss": 3.1685, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "grad_norm": 0.2794119715690613, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 3.2021, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "grad_norm": 0.27362996339797974, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 3.1814, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "grad_norm": 0.2530471384525299, | |
| "learning_rate": 2.4e-05, | |
| "loss": 3.2072, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 0.32529401779174805, | |
| "learning_rate": 2e-05, | |
| "loss": 3.155, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "grad_norm": 0.28048309683799744, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 3.1852, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "grad_norm": 0.3107221722602844, | |
| "learning_rate": 1.2e-05, | |
| "loss": 3.1617, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 0.2563643753528595, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 3.1996, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "grad_norm": 0.40821021795272827, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 3.1079, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 0.42536577582359314, | |
| "learning_rate": 0.0, | |
| "loss": 3.0919, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_clap": 0.09740716218948364, | |
| "eval_loss": 4.961584091186523, | |
| "eval_runtime": 50.0977, | |
| "eval_samples_per_second": 0.04, | |
| "eval_steps_per_second": 0.04, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 50, | |
| "total_flos": 22172758761600.0, | |
| "train_loss": 3.4368285846710207, | |
| "train_runtime": 280.1233, | |
| "train_samples_per_second": 1.428, | |
| "train_steps_per_second": 0.178 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 22172758761600.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |