| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.7199999999999998, | |
| "eval_steps": 25, | |
| "global_step": 48, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 90.5236587524414, | |
| "learning_rate": 0.0002, | |
| "loss": 9.8877, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 60.02783966064453, | |
| "learning_rate": 0.0002, | |
| "loss": 9.4404, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 102.20321655273438, | |
| "learning_rate": 0.0002, | |
| "loss": 8.8638, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 82.49861145019531, | |
| "learning_rate": 0.0002, | |
| "loss": 8.0872, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 81.60125732421875, | |
| "learning_rate": 0.0002, | |
| "loss": 7.2292, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 52.2266960144043, | |
| "learning_rate": 0.0002, | |
| "loss": 7.0472, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 75.96226501464844, | |
| "learning_rate": 0.0002, | |
| "loss": 7.2919, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 70.42675018310547, | |
| "learning_rate": 0.0002, | |
| "loss": 7.1477, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 52.09629821777344, | |
| "learning_rate": 0.0002, | |
| "loss": 6.899, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 28.537267684936523, | |
| "learning_rate": 0.0002, | |
| "loss": 6.7016, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 29.203533172607422, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1981, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 23.664485931396484, | |
| "learning_rate": 0.0002, | |
| "loss": 6.3924, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 9.723264694213867, | |
| "learning_rate": 0.0002, | |
| "loss": 3.2305, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 18.88973045349121, | |
| "learning_rate": 0.0002, | |
| "loss": 6.3592, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 20.770938873291016, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1203, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 10.570067405700684, | |
| "learning_rate": 0.0002, | |
| "loss": 6.4217, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 13.833821296691895, | |
| "learning_rate": 0.0002, | |
| "loss": 6.439, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 19.160987854003906, | |
| "learning_rate": 0.0002, | |
| "loss": 5.8242, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 10.99656867980957, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1287, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 16.58768081665039, | |
| "learning_rate": 0.0002, | |
| "loss": 6.408, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 11.93716812133789, | |
| "learning_rate": 0.0002, | |
| "loss": 6.3103, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 9.053961753845215, | |
| "learning_rate": 0.0002, | |
| "loss": 6.2692, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 19.67817497253418, | |
| "learning_rate": 0.0002, | |
| "loss": 5.9086, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 10.540741920471191, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1991, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 8.84691333770752, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1812, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_clap": 0.13957726955413818, | |
| "eval_loss": 5.896897315979004, | |
| "eval_runtime": 206.5396, | |
| "eval_samples_per_second": 0.155, | |
| "eval_steps_per_second": 0.155, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.9374849796295166, | |
| "learning_rate": 0.0002, | |
| "loss": 3.1408, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 6.420533180236816, | |
| "learning_rate": 0.0002, | |
| "loss": 6.196, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 10.713640213012695, | |
| "learning_rate": 0.0002, | |
| "loss": 6.4188, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 13.720645904541016, | |
| "learning_rate": 0.0002, | |
| "loss": 5.9628, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 13.317567825317383, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1474, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 11.3407621383667, | |
| "learning_rate": 0.0002, | |
| "loss": 5.9697, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 9.393121719360352, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1152, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 14.54958438873291, | |
| "learning_rate": 0.0002, | |
| "loss": 5.7045, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 13.25341796875, | |
| "learning_rate": 0.0002, | |
| "loss": 6.2739, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 10.817360877990723, | |
| "learning_rate": 0.0002, | |
| "loss": 5.9946, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 18.011138916015625, | |
| "learning_rate": 0.0002, | |
| "loss": 5.5463, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 18.857688903808594, | |
| "learning_rate": 0.0002, | |
| "loss": 6.4324, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 10.496877670288086, | |
| "learning_rate": 0.0002, | |
| "loss": 5.9915, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.552528381347656, | |
| "learning_rate": 0.0002, | |
| "loss": 3.1022, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 9.081900596618652, | |
| "learning_rate": 0.0002, | |
| "loss": 6.0394, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 13.829048156738281, | |
| "learning_rate": 0.0002, | |
| "loss": 5.9436, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "grad_norm": 6.3993120193481445, | |
| "learning_rate": 0.0002, | |
| "loss": 6.276, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 20.436588287353516, | |
| "learning_rate": 0.0002, | |
| "loss": 5.7975, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 7.475179672241211, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1817, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "grad_norm": 17.927181243896484, | |
| "learning_rate": 0.0002, | |
| "loss": 5.8574, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "grad_norm": 18.49729347229004, | |
| "learning_rate": 0.0002, | |
| "loss": 5.4645, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 16.93431854248047, | |
| "learning_rate": 0.0002, | |
| "loss": 5.7863, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 3.7199999999999998, | |
| "grad_norm": 10.554439544677734, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1658, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.7199999999999998, | |
| "step": 48, | |
| "total_flos": 180716776957944.0, | |
| "train_loss": 6.281136025985082, | |
| "train_runtime": 1764.029, | |
| "train_samples_per_second": 0.907, | |
| "train_steps_per_second": 0.027 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 48, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 180716776957944.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |