| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "eval_steps": 1000, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 13.470948224931323, | |
| "learning_rate": 5e-06, | |
| "loss": 2.9758, | |
| "mean_token_accuracy": 0.6230694115161896, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 13.750445042685815, | |
| "learning_rate": 1e-05, | |
| "loss": 2.659, | |
| "mean_token_accuracy": 0.6543959975242615, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 2.698895165960937, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 2.1537, | |
| "mean_token_accuracy": 0.7085180535912514, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.021193946571878, | |
| "learning_rate": 2e-05, | |
| "loss": 1.7625, | |
| "mean_token_accuracy": 0.7400283828377724, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 1.8876805703856514, | |
| "learning_rate": 1.9961946980917457e-05, | |
| "loss": 1.4332, | |
| "mean_token_accuracy": 0.7722540438175202, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.9137993288001238, | |
| "learning_rate": 1.9848077530122083e-05, | |
| "loss": 1.1868, | |
| "mean_token_accuracy": 0.7999878883361816, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "grad_norm": 1.864430389242646, | |
| "learning_rate": 1.9659258262890683e-05, | |
| "loss": 0.9777, | |
| "mean_token_accuracy": 0.8262525200843811, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.8647110420709154, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.7848, | |
| "mean_token_accuracy": 0.8553504809737206, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "grad_norm": 1.8688016725450207, | |
| "learning_rate": 1.9063077870366504e-05, | |
| "loss": 0.5982, | |
| "mean_token_accuracy": 0.8966465726494789, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 1.9284394076191518, | |
| "learning_rate": 1.866025403784439e-05, | |
| "loss": 0.4319, | |
| "mean_token_accuracy": 0.9422784462571144, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "grad_norm": 1.9014044414549942, | |
| "learning_rate": 1.819152044288992e-05, | |
| "loss": 0.3002, | |
| "mean_token_accuracy": 0.9657423257827759, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 1.837847610581908, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.2033, | |
| "mean_token_accuracy": 0.9815968632698059, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 32.5, | |
| "grad_norm": 1.7961955203879951, | |
| "learning_rate": 1.7071067811865477e-05, | |
| "loss": 0.1403, | |
| "mean_token_accuracy": 0.9899852931499481, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 1.7767287140174859, | |
| "learning_rate": 1.6427876096865394e-05, | |
| "loss": 0.1046, | |
| "mean_token_accuracy": 0.9929877564311027, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 37.5, | |
| "grad_norm": 1.4092119906646094, | |
| "learning_rate": 1.573576436351046e-05, | |
| "loss": 0.0739, | |
| "mean_token_accuracy": 0.9973135724663734, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 1.1487581792063775, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.0542, | |
| "mean_token_accuracy": 0.9987073212862014, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 42.5, | |
| "grad_norm": 0.9136680911828254, | |
| "learning_rate": 1.4226182617406996e-05, | |
| "loss": 0.0423, | |
| "mean_token_accuracy": 0.9989680975675583, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 0.714276120050603, | |
| "learning_rate": 1.342020143325669e-05, | |
| "loss": 0.0342, | |
| "mean_token_accuracy": 0.9991405799984932, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 47.5, | |
| "grad_norm": 0.601809233243359, | |
| "learning_rate": 1.2588190451025209e-05, | |
| "loss": 0.0285, | |
| "mean_token_accuracy": 0.9993251755833625, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 0.5529472873720432, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.0244, | |
| "mean_token_accuracy": 0.9994255259633065, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 52.5, | |
| "grad_norm": 0.522508977878471, | |
| "learning_rate": 1.0871557427476585e-05, | |
| "loss": 0.0213, | |
| "mean_token_accuracy": 0.9994575038552285, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 0.5010419801089824, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0189, | |
| "mean_token_accuracy": 0.9994710594415664, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 57.5, | |
| "grad_norm": 0.4737936422226576, | |
| "learning_rate": 9.128442572523418e-06, | |
| "loss": 0.0169, | |
| "mean_token_accuracy": 0.9995683461427689, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 0.45159464312393605, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.0155, | |
| "mean_token_accuracy": 0.9995812460780144, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 62.5, | |
| "grad_norm": 0.4315104257213208, | |
| "learning_rate": 7.411809548974792e-06, | |
| "loss": 0.0141, | |
| "mean_token_accuracy": 0.9995992064476014, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 0.41026001713760674, | |
| "learning_rate": 6.579798566743314e-06, | |
| "loss": 0.0131, | |
| "mean_token_accuracy": 0.9995993703603745, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 67.5, | |
| "grad_norm": 0.39156974708559894, | |
| "learning_rate": 5.773817382593008e-06, | |
| "loss": 0.0122, | |
| "mean_token_accuracy": 0.9996095344424247, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 0.3758286740687803, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.0116, | |
| "mean_token_accuracy": 0.9996033638715744, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 72.5, | |
| "grad_norm": 0.36650688039325785, | |
| "learning_rate": 4.264235636489542e-06, | |
| "loss": 0.011, | |
| "mean_token_accuracy": 0.9996233299374581, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 0.355312570554161, | |
| "learning_rate": 3.5721239031346067e-06, | |
| "loss": 0.0107, | |
| "mean_token_accuracy": 0.9996237218379974, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 77.5, | |
| "grad_norm": 0.3470766956761054, | |
| "learning_rate": 2.9289321881345257e-06, | |
| "loss": 0.0102, | |
| "mean_token_accuracy": 0.9996343672275543, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 0.3405501982761911, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.0101, | |
| "mean_token_accuracy": 0.9996294215321541, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 82.5, | |
| "grad_norm": 0.3355965449228851, | |
| "learning_rate": 1.808479557110081e-06, | |
| "loss": 0.0098, | |
| "mean_token_accuracy": 0.9996438756585121, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 0.3310960741356627, | |
| "learning_rate": 1.339745962155613e-06, | |
| "loss": 0.0096, | |
| "mean_token_accuracy": 0.9996403336524964, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 87.5, | |
| "grad_norm": 0.3296565871751418, | |
| "learning_rate": 9.369221296335007e-07, | |
| "loss": 0.0096, | |
| "mean_token_accuracy": 0.9996464192867279, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 0.32579571171147914, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.0095, | |
| "mean_token_accuracy": 0.9996390745043755, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 92.5, | |
| "grad_norm": 0.3268532655575783, | |
| "learning_rate": 3.4074173710931804e-07, | |
| "loss": 0.0095, | |
| "mean_token_accuracy": 0.9996397033333778, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 0.32376162339876563, | |
| "learning_rate": 1.519224698779198e-07, | |
| "loss": 0.0093, | |
| "mean_token_accuracy": 0.9996532008051873, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 97.5, | |
| "grad_norm": 0.3250121331654675, | |
| "learning_rate": 3.805301908254455e-08, | |
| "loss": 0.0094, | |
| "mean_token_accuracy": 0.9996466800570488, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 0.3261372439108348, | |
| "learning_rate": 0.0, | |
| "loss": 0.0093, | |
| "mean_token_accuracy": 0.999645346403122, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 34146710650880.0, | |
| "train_loss": 0.40527356846258045, | |
| "train_runtime": 2669.3514, | |
| "train_samples_per_second": 3.559, | |
| "train_steps_per_second": 0.075 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 34146710650880.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |