| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9955555555555555, | |
| "eval_steps": 500, | |
| "global_step": 161, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.006183574879227053, | |
| "grad_norm": 6.14696354757804, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3449, | |
| "mean_token_accuracy": 0.915102468803525, | |
| "num_tokens": 536845.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06183574879227053, | |
| "grad_norm": 0.25824490458409005, | |
| "learning_rate": 0.0001, | |
| "loss": 0.255, | |
| "mean_token_accuracy": 0.91935970261693, | |
| "num_tokens": 5420782.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12367149758454106, | |
| "grad_norm": 0.12207035150262055, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1607, | |
| "mean_token_accuracy": 0.9383567951619625, | |
| "num_tokens": 10836749.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1855072463768116, | |
| "grad_norm": 0.0931360663742762, | |
| "learning_rate": 0.0001, | |
| "loss": 0.14, | |
| "mean_token_accuracy": 0.945984673500061, | |
| "num_tokens": 16250383.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.24734299516908212, | |
| "grad_norm": 0.08260645299611312, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1294, | |
| "mean_token_accuracy": 0.9498019333928823, | |
| "num_tokens": 21694294.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.30917874396135264, | |
| "grad_norm": 0.09270504271704304, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1231, | |
| "mean_token_accuracy": 0.952541628666222, | |
| "num_tokens": 27089618.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3710144927536232, | |
| "grad_norm": 0.09892330366085841, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1173, | |
| "mean_token_accuracy": 0.9548029892146588, | |
| "num_tokens": 32497984.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.43285024154589374, | |
| "grad_norm": 0.09081855122666112, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1129, | |
| "mean_token_accuracy": 0.9565541172400117, | |
| "num_tokens": 37913108.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.49468599033816424, | |
| "grad_norm": 0.08744357034359414, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1093, | |
| "mean_token_accuracy": 0.9580830905586482, | |
| "num_tokens": 43329106.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5565217391304348, | |
| "grad_norm": 0.09900086305146383, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1064, | |
| "mean_token_accuracy": 0.9591556312516332, | |
| "num_tokens": 48742706.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6183574879227053, | |
| "grad_norm": 0.10897858928578598, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1047, | |
| "mean_token_accuracy": 0.9600882643833757, | |
| "num_tokens": 54145465.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6801932367149759, | |
| "grad_norm": 0.10086205886710188, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1012, | |
| "mean_token_accuracy": 0.9615297164767981, | |
| "num_tokens": 59564822.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7420289855072464, | |
| "grad_norm": 0.09201974209736791, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0995, | |
| "mean_token_accuracy": 0.9620782939717174, | |
| "num_tokens": 64974349.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8038647342995169, | |
| "grad_norm": 0.10194015491082806, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0981, | |
| "mean_token_accuracy": 0.9625154824927449, | |
| "num_tokens": 70366254.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.8657004830917875, | |
| "grad_norm": 0.09022409759364422, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0972, | |
| "mean_token_accuracy": 0.9628898801282049, | |
| "num_tokens": 75762074.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.927536231884058, | |
| "grad_norm": 0.08192921340517849, | |
| "learning_rate": 0.0001, | |
| "loss": 0.095, | |
| "mean_token_accuracy": 0.963659605011344, | |
| "num_tokens": 81167608.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.9893719806763285, | |
| "grad_norm": 0.08255379291969749, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0943, | |
| "mean_token_accuracy": 0.9640055214986205, | |
| "num_tokens": 86563355.0, | |
| "step": 160 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 644, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 233965155254272.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |