| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0010260619741432383, | |
| "eval_steps": 4, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.420206580477461e-05, | |
| "grad_norm": 3.4161529541015625, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 7.5749, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 3.420206580477461e-05, | |
| "eval_loss": 3.9098658561706543, | |
| "eval_runtime": 2566.1323, | |
| "eval_samples_per_second": 4.797, | |
| "eval_steps_per_second": 1.199, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 6.840413160954922e-05, | |
| "grad_norm": 3.0953049659729004, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 6.0707, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00010260619741432382, | |
| "grad_norm": 3.0091123580932617, | |
| "learning_rate": 6e-06, | |
| "loss": 7.0997, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00013680826321909844, | |
| "grad_norm": 3.4014148712158203, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 7.5191, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00013680826321909844, | |
| "eval_loss": 3.9077231884002686, | |
| "eval_runtime": 2567.2782, | |
| "eval_samples_per_second": 4.795, | |
| "eval_steps_per_second": 1.199, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00017101032902387304, | |
| "grad_norm": 3.0993244647979736, | |
| "learning_rate": 1e-05, | |
| "loss": 7.0823, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00020521239482864764, | |
| "grad_norm": 4.345460414886475, | |
| "learning_rate": 9.960573506572391e-06, | |
| "loss": 9.6282, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00023941446063342227, | |
| "grad_norm": 3.2689383029937744, | |
| "learning_rate": 9.842915805643156e-06, | |
| "loss": 7.1881, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00027361652643819687, | |
| "grad_norm": 3.5341312885284424, | |
| "learning_rate": 9.648882429441258e-06, | |
| "loss": 6.6592, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00027361652643819687, | |
| "eval_loss": 3.8825018405914307, | |
| "eval_runtime": 2566.2011, | |
| "eval_samples_per_second": 4.797, | |
| "eval_steps_per_second": 1.199, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0003078185922429715, | |
| "grad_norm": 4.909244537353516, | |
| "learning_rate": 9.381533400219319e-06, | |
| "loss": 9.9852, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0003420206580477461, | |
| "grad_norm": 3.6594748497009277, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 7.8606, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0003762227238525207, | |
| "grad_norm": 5.5325164794921875, | |
| "learning_rate": 8.644843137107058e-06, | |
| "loss": 10.5745, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0004104247896572953, | |
| "grad_norm": 3.811521053314209, | |
| "learning_rate": 8.18711994874345e-06, | |
| "loss": 6.9418, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0004104247896572953, | |
| "eval_loss": 3.8427975177764893, | |
| "eval_runtime": 2566.033, | |
| "eval_samples_per_second": 4.798, | |
| "eval_steps_per_second": 1.2, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0004446268554620699, | |
| "grad_norm": 4.672744274139404, | |
| "learning_rate": 7.679133974894984e-06, | |
| "loss": 8.3155, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00047882892126684454, | |
| "grad_norm": 4.548338413238525, | |
| "learning_rate": 7.128896457825364e-06, | |
| "loss": 9.1416, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0005130309870716191, | |
| "grad_norm": 4.175900936126709, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 7.591, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0005472330528763937, | |
| "grad_norm": 3.9903697967529297, | |
| "learning_rate": 5.936906572928625e-06, | |
| "loss": 7.0457, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0005472330528763937, | |
| "eval_loss": 3.7900733947753906, | |
| "eval_runtime": 2565.8582, | |
| "eval_samples_per_second": 4.798, | |
| "eval_steps_per_second": 1.2, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0005814351186811683, | |
| "grad_norm": 4.355044841766357, | |
| "learning_rate": 5.3139525976465675e-06, | |
| "loss": 6.9375, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.000615637184485943, | |
| "grad_norm": 6.346569538116455, | |
| "learning_rate": 4.686047402353433e-06, | |
| "loss": 10.2759, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0006498392502907176, | |
| "grad_norm": 8.13481616973877, | |
| "learning_rate": 4.063093427071376e-06, | |
| "loss": 11.3094, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0006840413160954922, | |
| "grad_norm": 4.394895553588867, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 6.8632, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0006840413160954922, | |
| "eval_loss": 3.7471675872802734, | |
| "eval_runtime": 2566.024, | |
| "eval_samples_per_second": 4.798, | |
| "eval_steps_per_second": 1.2, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0007182433819002668, | |
| "grad_norm": 5.748673439025879, | |
| "learning_rate": 2.871103542174637e-06, | |
| "loss": 9.2619, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0007524454477050414, | |
| "grad_norm": 5.746485233306885, | |
| "learning_rate": 2.320866025105016e-06, | |
| "loss": 8.7318, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.000786647513509816, | |
| "grad_norm": 4.3517255783081055, | |
| "learning_rate": 1.8128800512565514e-06, | |
| "loss": 6.9672, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0008208495793145906, | |
| "grad_norm": 4.058344841003418, | |
| "learning_rate": 1.3551568628929434e-06, | |
| "loss": 7.4406, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0008208495793145906, | |
| "eval_loss": 3.7231340408325195, | |
| "eval_runtime": 2566.6279, | |
| "eval_samples_per_second": 4.797, | |
| "eval_steps_per_second": 1.199, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0008550516451193652, | |
| "grad_norm": 5.805904865264893, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 7.5462, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0008892537109241398, | |
| "grad_norm": 5.130537986755371, | |
| "learning_rate": 6.184665997806832e-07, | |
| "loss": 7.2794, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0009234557767289145, | |
| "grad_norm": 5.133691310882568, | |
| "learning_rate": 3.511175705587433e-07, | |
| "loss": 7.9919, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0009576578425336891, | |
| "grad_norm": 5.328059196472168, | |
| "learning_rate": 1.5708419435684463e-07, | |
| "loss": 7.8922, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0009576578425336891, | |
| "eval_loss": 3.712865114212036, | |
| "eval_runtime": 2567.0389, | |
| "eval_samples_per_second": 4.796, | |
| "eval_steps_per_second": 1.199, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0009918599083384636, | |
| "grad_norm": 5.650175094604492, | |
| "learning_rate": 3.9426493427611177e-08, | |
| "loss": 8.2262, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0010260619741432383, | |
| "grad_norm": 4.0064263343811035, | |
| "learning_rate": 0.0, | |
| "loss": 6.046, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 20, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.90434898870272e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |