| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 30, | |
| "global_step": 192, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010452961672473868, | |
| "grad_norm": 1.415148138999939, | |
| "learning_rate": 0.0, | |
| "loss": 0.737, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05226480836236934, | |
| "grad_norm": 1.0576598644256592, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.6984, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.10452961672473868, | |
| "grad_norm": 0.21559220552444458, | |
| "learning_rate": 9.993582535855263e-05, | |
| "loss": 0.5093, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.156794425087108, | |
| "grad_norm": 0.20353108644485474, | |
| "learning_rate": 9.954424340791196e-05, | |
| "loss": 0.4536, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.20905923344947736, | |
| "grad_norm": 0.3409803807735443, | |
| "learning_rate": 9.879951981385578e-05, | |
| "loss": 0.4662, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2613240418118467, | |
| "grad_norm": 0.11676070839166641, | |
| "learning_rate": 9.770696282000244e-05, | |
| "loss": 0.4243, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.313588850174216, | |
| "grad_norm": 0.09078264981508255, | |
| "learning_rate": 9.627435995817799e-05, | |
| "loss": 0.2723, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.313588850174216, | |
| "eval_loss": 0.31834542751312256, | |
| "eval_runtime": 1748.9904, | |
| "eval_samples_per_second": 0.292, | |
| "eval_steps_per_second": 0.073, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.36585365853658536, | |
| "grad_norm": 0.08448098599910736, | |
| "learning_rate": 9.451192254041758e-05, | |
| "loss": 0.285, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.4181184668989547, | |
| "grad_norm": 0.0991150364279747, | |
| "learning_rate": 9.243221287473756e-05, | |
| "loss": 0.3197, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.47038327526132406, | |
| "grad_norm": 0.17393162846565247, | |
| "learning_rate": 9.005005472346924e-05, | |
| "loss": 0.3749, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5226480836236934, | |
| "grad_norm": 0.0953838899731636, | |
| "learning_rate": 8.738242764239046e-05, | |
| "loss": 0.2699, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5749128919860628, | |
| "grad_norm": 0.07958400994539261, | |
| "learning_rate": 8.444834595378434e-05, | |
| "loss": 0.2421, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.627177700348432, | |
| "grad_norm": 0.08708363026380539, | |
| "learning_rate": 8.126872321608184e-05, | |
| "loss": 0.2645, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.627177700348432, | |
| "eval_loss": 0.26565828919410706, | |
| "eval_runtime": 1749.0443, | |
| "eval_samples_per_second": 0.292, | |
| "eval_steps_per_second": 0.073, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6794425087108014, | |
| "grad_norm": 0.10611709207296371, | |
| "learning_rate": 7.786622315612183e-05, | |
| "loss": 0.3069, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.7317073170731707, | |
| "grad_norm": 0.093358114361763, | |
| "learning_rate": 7.426509812655406e-05, | |
| "loss": 0.3514, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.7839721254355401, | |
| "grad_norm": 0.0683087483048439, | |
| "learning_rate": 7.049101623982937e-05, | |
| "loss": 0.2057, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.8362369337979094, | |
| "grad_norm": 0.0792144313454628, | |
| "learning_rate": 6.65708784109318e-05, | |
| "loss": 0.2101, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.8885017421602788, | |
| "grad_norm": 0.09680237621068954, | |
| "learning_rate": 6.253262661293604e-05, | |
| "loss": 0.2638, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.9407665505226481, | |
| "grad_norm": 0.1257801651954651, | |
| "learning_rate": 5.840504471210742e-05, | |
| "loss": 0.3075, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.9407665505226481, | |
| "eval_loss": 0.2471843957901001, | |
| "eval_runtime": 1748.4786, | |
| "eval_samples_per_second": 0.292, | |
| "eval_steps_per_second": 0.073, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.9930313588850174, | |
| "grad_norm": 0.10834779590368271, | |
| "learning_rate": 5.4217553302152237e-05, | |
| "loss": 0.2944, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.0418118466898956, | |
| "grad_norm": 0.0633026584982872, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2095, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0940766550522647, | |
| "grad_norm": 0.07635607570409775, | |
| "learning_rate": 4.578244669784777e-05, | |
| "loss": 0.2058, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.146341463414634, | |
| "grad_norm": 0.0802493542432785, | |
| "learning_rate": 4.15949552878926e-05, | |
| "loss": 0.2232, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.1986062717770034, | |
| "grad_norm": 0.12653642892837524, | |
| "learning_rate": 3.746737338706397e-05, | |
| "loss": 0.2532, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.2508710801393728, | |
| "grad_norm": 0.07813160121440887, | |
| "learning_rate": 3.3429121589068215e-05, | |
| "loss": 0.2893, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.2508710801393728, | |
| "eval_loss": 0.239236518740654, | |
| "eval_runtime": 1751.4721, | |
| "eval_samples_per_second": 0.291, | |
| "eval_steps_per_second": 0.073, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3031358885017421, | |
| "grad_norm": 0.07795720547437668, | |
| "learning_rate": 2.950898376017064e-05, | |
| "loss": 0.1842, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.3554006968641115, | |
| "grad_norm": 0.07542526721954346, | |
| "learning_rate": 2.573490187344596e-05, | |
| "loss": 0.2031, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4076655052264808, | |
| "grad_norm": 0.10047340393066406, | |
| "learning_rate": 2.2133776843878186e-05, | |
| "loss": 0.24, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.4599303135888502, | |
| "grad_norm": 0.13595731556415558, | |
| "learning_rate": 1.873127678391816e-05, | |
| "loss": 0.2808, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5121951219512195, | |
| "grad_norm": 0.06210995092988014, | |
| "learning_rate": 1.555165404621567e-05, | |
| "loss": 0.235, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.5644599303135889, | |
| "grad_norm": 0.08401988446712494, | |
| "learning_rate": 1.2617572357609564e-05, | |
| "loss": 0.1849, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.5644599303135889, | |
| "eval_loss": 0.23435795307159424, | |
| "eval_runtime": 1753.006, | |
| "eval_samples_per_second": 0.291, | |
| "eval_steps_per_second": 0.073, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6167247386759582, | |
| "grad_norm": 0.07571443915367126, | |
| "learning_rate": 9.949945276530781e-06, | |
| "loss": 0.205, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.6689895470383276, | |
| "grad_norm": 0.08948186039924622, | |
| "learning_rate": 7.5677871252624485e-06, | |
| "loss": 0.2501, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.721254355400697, | |
| "grad_norm": 0.185760036110878, | |
| "learning_rate": 5.488077459582425e-06, | |
| "loss": 0.3175, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.773519163763066, | |
| "grad_norm": 0.055869363248348236, | |
| "learning_rate": 3.7256400418220262e-06, | |
| "loss": 0.1723, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8257839721254356, | |
| "grad_norm": 0.0660533756017685, | |
| "learning_rate": 2.2930371799975594e-06, | |
| "loss": 0.1959, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.8780487804878048, | |
| "grad_norm": 0.07585973292589188, | |
| "learning_rate": 1.2004801861442371e-06, | |
| "loss": 0.2145, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.8780487804878048, | |
| "eval_loss": 0.23282098770141602, | |
| "eval_runtime": 1752.5559, | |
| "eval_samples_per_second": 0.291, | |
| "eval_steps_per_second": 0.073, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9303135888501743, | |
| "grad_norm": 0.10959440469741821, | |
| "learning_rate": 4.55756592088058e-07, | |
| "loss": 0.2618, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.9825783972125435, | |
| "grad_norm": 0.07039328664541245, | |
| "learning_rate": 6.417464144736208e-08, | |
| "loss": 0.2559, | |
| "step": 190 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 192, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 30, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2340088273817969e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |