| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 200, | |
| "global_step": 422, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05924170616113744, | |
| "grad_norm": 2.457188844680786, | |
| "learning_rate": 5.8139534883720935e-06, | |
| "loss": 11.6813, | |
| "mean_token_accuracy": 1.2189515546197072e-05, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.11848341232227488, | |
| "grad_norm": 2.1204864978790283, | |
| "learning_rate": 9.815303430079157e-06, | |
| "loss": 11.6667, | |
| "mean_token_accuracy": 2.3536479129688815e-05, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17772511848341233, | |
| "grad_norm": 2.153775453567505, | |
| "learning_rate": 9.155672823218998e-06, | |
| "loss": 11.6042, | |
| "mean_token_accuracy": 5.270051406114362e-05, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.23696682464454977, | |
| "grad_norm": 2.3368475437164307, | |
| "learning_rate": 8.49604221635884e-06, | |
| "loss": 11.4981, | |
| "mean_token_accuracy": 0.0005879536968132015, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2962085308056872, | |
| "grad_norm": 2.508676052093506, | |
| "learning_rate": 7.836411609498681e-06, | |
| "loss": 11.3824, | |
| "mean_token_accuracy": 0.005088501239588368, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.35545023696682465, | |
| "grad_norm": 2.4591517448425293, | |
| "learning_rate": 7.176781002638523e-06, | |
| "loss": 11.2577, | |
| "mean_token_accuracy": 0.018929289733059705, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4146919431279621, | |
| "grad_norm": 2.546363115310669, | |
| "learning_rate": 6.517150395778365e-06, | |
| "loss": 11.11, | |
| "mean_token_accuracy": 0.03834218932315707, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.47393364928909953, | |
| "grad_norm": 2.765162944793701, | |
| "learning_rate": 5.857519788918207e-06, | |
| "loss": 10.9715, | |
| "mean_token_accuracy": 0.04483937283977866, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.47393364928909953, | |
| "eval_loss": 10.91100025177002, | |
| "eval_mean_token_accuracy": 0.04610209306702018, | |
| "eval_runtime": 19.5293, | |
| "eval_samples_per_second": 153.615, | |
| "eval_steps_per_second": 9.627, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.533175355450237, | |
| "grad_norm": 2.6922924518585205, | |
| "learning_rate": 5.197889182058048e-06, | |
| "loss": 10.8376, | |
| "mean_token_accuracy": 0.046522868759930135, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.5924170616113744, | |
| "grad_norm": 2.660522222518921, | |
| "learning_rate": 4.538258575197889e-06, | |
| "loss": 10.7329, | |
| "mean_token_accuracy": 0.046305897459387776, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.6516587677725119, | |
| "grad_norm": 2.878413200378418, | |
| "learning_rate": 3.878627968337731e-06, | |
| "loss": 10.6505, | |
| "mean_token_accuracy": 0.04519957359880209, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.7109004739336493, | |
| "grad_norm": 2.853266954421997, | |
| "learning_rate": 3.2189973614775725e-06, | |
| "loss": 10.5696, | |
| "mean_token_accuracy": 0.04625188913196325, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7701421800947867, | |
| "grad_norm": 2.903547525405884, | |
| "learning_rate": 2.5593667546174145e-06, | |
| "loss": 10.5126, | |
| "mean_token_accuracy": 0.045716633237898346, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.8293838862559242, | |
| "grad_norm": 2.624979257583618, | |
| "learning_rate": 1.8997361477572559e-06, | |
| "loss": 10.4573, | |
| "mean_token_accuracy": 0.04610322181135416, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.8886255924170616, | |
| "grad_norm": 2.7634570598602295, | |
| "learning_rate": 1.2401055408970976e-06, | |
| "loss": 10.43, | |
| "mean_token_accuracy": 0.046120549254119396, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.9478672985781991, | |
| "grad_norm": 2.7394840717315674, | |
| "learning_rate": 5.804749340369393e-07, | |
| "loss": 10.4171, | |
| "mean_token_accuracy": 0.046127126961946485, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.9478672985781991, | |
| "eval_loss": 10.411770820617676, | |
| "eval_mean_token_accuracy": 0.04626982772049117, | |
| "eval_runtime": 19.5552, | |
| "eval_samples_per_second": 153.412, | |
| "eval_steps_per_second": 9.614, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 422, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.1954772000768e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |