| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00641094133988674, | |
| "eval_steps": 4, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00021369804466289135, | |
| "grad_norm": 2.0482637882232666, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 5.0098, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00021369804466289135, | |
| "eval_loss": 2.4674696922302246, | |
| "eval_runtime": 411.0943, | |
| "eval_samples_per_second": 4.795, | |
| "eval_steps_per_second": 1.199, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0004273960893257827, | |
| "grad_norm": 2.238199234008789, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 5.2527, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.000641094133988674, | |
| "grad_norm": 1.965638518333435, | |
| "learning_rate": 6e-06, | |
| "loss": 4.9741, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0008547921786515654, | |
| "grad_norm": 2.046441078186035, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 5.1034, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0008547921786515654, | |
| "eval_loss": 2.4658350944519043, | |
| "eval_runtime": 411.1677, | |
| "eval_samples_per_second": 4.794, | |
| "eval_steps_per_second": 1.199, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0010684902233144566, | |
| "grad_norm": 2.407226085662842, | |
| "learning_rate": 1e-05, | |
| "loss": 4.818, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.001282188267977348, | |
| "grad_norm": 2.8345041275024414, | |
| "learning_rate": 9.960573506572391e-06, | |
| "loss": 5.3071, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0014958863126402393, | |
| "grad_norm": 2.036144256591797, | |
| "learning_rate": 9.842915805643156e-06, | |
| "loss": 4.7743, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0017095843573031308, | |
| "grad_norm": 2.0334997177124023, | |
| "learning_rate": 9.648882429441258e-06, | |
| "loss": 4.9792, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0017095843573031308, | |
| "eval_loss": 2.4481966495513916, | |
| "eval_runtime": 411.0364, | |
| "eval_samples_per_second": 4.795, | |
| "eval_steps_per_second": 1.199, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.001923282401966022, | |
| "grad_norm": 1.955418586730957, | |
| "learning_rate": 9.381533400219319e-06, | |
| "loss": 4.6347, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0021369804466289132, | |
| "grad_norm": 2.774395704269409, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 5.2668, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0023506784912918047, | |
| "grad_norm": 2.4080920219421387, | |
| "learning_rate": 8.644843137107058e-06, | |
| "loss": 4.669, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.002564376535954696, | |
| "grad_norm": 2.503000020980835, | |
| "learning_rate": 8.18711994874345e-06, | |
| "loss": 4.6545, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.002564376535954696, | |
| "eval_loss": 2.4184906482696533, | |
| "eval_runtime": 410.9238, | |
| "eval_samples_per_second": 4.797, | |
| "eval_steps_per_second": 1.2, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.002778074580617587, | |
| "grad_norm": 2.407099485397339, | |
| "learning_rate": 7.679133974894984e-06, | |
| "loss": 5.1754, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0029917726252804786, | |
| "grad_norm": 2.5891873836517334, | |
| "learning_rate": 7.128896457825364e-06, | |
| "loss": 4.9656, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.00320547066994337, | |
| "grad_norm": 2.612175464630127, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 4.9479, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0034191687146062615, | |
| "grad_norm": 2.592885732650757, | |
| "learning_rate": 5.936906572928625e-06, | |
| "loss": 4.8353, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0034191687146062615, | |
| "eval_loss": 2.3783442974090576, | |
| "eval_runtime": 411.1282, | |
| "eval_samples_per_second": 4.794, | |
| "eval_steps_per_second": 1.199, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0036328667592691525, | |
| "grad_norm": 2.204364538192749, | |
| "learning_rate": 5.3139525976465675e-06, | |
| "loss": 4.4296, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.003846564803932044, | |
| "grad_norm": 2.751991033554077, | |
| "learning_rate": 4.686047402353433e-06, | |
| "loss": 4.6077, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.004060262848594935, | |
| "grad_norm": 2.8248343467712402, | |
| "learning_rate": 4.063093427071376e-06, | |
| "loss": 4.8522, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0042739608932578265, | |
| "grad_norm": 2.5259201526641846, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 4.4364, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0042739608932578265, | |
| "eval_loss": 2.3454465866088867, | |
| "eval_runtime": 411.3946, | |
| "eval_samples_per_second": 4.791, | |
| "eval_steps_per_second": 1.198, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.004487658937920718, | |
| "grad_norm": 2.3242313861846924, | |
| "learning_rate": 2.871103542174637e-06, | |
| "loss": 4.729, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.004701356982583609, | |
| "grad_norm": 2.1803932189941406, | |
| "learning_rate": 2.320866025105016e-06, | |
| "loss": 4.4493, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.004915055027246501, | |
| "grad_norm": 3.0520858764648438, | |
| "learning_rate": 1.8128800512565514e-06, | |
| "loss": 4.7408, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.005128753071909392, | |
| "grad_norm": 3.3410630226135254, | |
| "learning_rate": 1.3551568628929434e-06, | |
| "loss": 4.6984, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.005128753071909392, | |
| "eval_loss": 2.328244924545288, | |
| "eval_runtime": 411.355, | |
| "eval_samples_per_second": 4.791, | |
| "eval_steps_per_second": 1.198, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.005342451116572284, | |
| "grad_norm": 2.358137845993042, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 4.4836, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.005556149161235174, | |
| "grad_norm": 2.9494125843048096, | |
| "learning_rate": 6.184665997806832e-07, | |
| "loss": 5.1003, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.005769847205898066, | |
| "grad_norm": 3.0665676593780518, | |
| "learning_rate": 3.511175705587433e-07, | |
| "loss": 4.7526, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.005983545250560957, | |
| "grad_norm": 2.681220293045044, | |
| "learning_rate": 1.5708419435684463e-07, | |
| "loss": 4.6398, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.005983545250560957, | |
| "eval_loss": 2.3202476501464844, | |
| "eval_runtime": 411.2537, | |
| "eval_samples_per_second": 4.793, | |
| "eval_steps_per_second": 1.199, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.006197243295223849, | |
| "grad_norm": 3.1423282623291016, | |
| "learning_rate": 3.9426493427611177e-08, | |
| "loss": 4.7534, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.00641094133988674, | |
| "grad_norm": 2.7747702598571777, | |
| "learning_rate": 0.0, | |
| "loss": 4.8573, | |
| "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 | |
| } | |