| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 100, | |
| "global_step": 169, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.029585798816568046, | |
| "grad_norm": 2.297424387110319, | |
| "learning_rate": 5.882352941176471e-06, | |
| "loss": 1.1002, | |
| "mean_token_accuracy": 0.710122095857212, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05917159763313609, | |
| "grad_norm": 1.5962459582070525, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 1.0326, | |
| "mean_token_accuracy": 0.7201974025955006, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08875739644970414, | |
| "grad_norm": 0.84813226022544, | |
| "learning_rate": 1.7647058823529414e-05, | |
| "loss": 0.9517, | |
| "mean_token_accuracy": 0.7326376870412972, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.11834319526627218, | |
| "grad_norm": 0.6449426089322932, | |
| "learning_rate": 1.9980782984658682e-05, | |
| "loss": 0.8804, | |
| "mean_token_accuracy": 0.7473364378856827, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.14792899408284024, | |
| "grad_norm": 0.5511541331468433, | |
| "learning_rate": 1.9863613034027224e-05, | |
| "loss": 0.8536, | |
| "mean_token_accuracy": 0.7518443039297401, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.17751479289940827, | |
| "grad_norm": 0.442628562609923, | |
| "learning_rate": 1.9641197940012136e-05, | |
| "loss": 0.8404, | |
| "mean_token_accuracy": 0.7545613836853171, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.20710059171597633, | |
| "grad_norm": 0.3950445105792925, | |
| "learning_rate": 1.9315910880512792e-05, | |
| "loss": 0.8212, | |
| "mean_token_accuracy": 0.7580693317680411, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.23668639053254437, | |
| "grad_norm": 0.38499101012568593, | |
| "learning_rate": 1.8891222681391853e-05, | |
| "loss": 0.805, | |
| "mean_token_accuracy": 0.7621003697400892, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.26627218934911245, | |
| "grad_norm": 0.426957308734079, | |
| "learning_rate": 1.8371664782625287e-05, | |
| "loss": 0.7984, | |
| "mean_token_accuracy": 0.763465327169843, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2958579881656805, | |
| "grad_norm": 0.3931886133887098, | |
| "learning_rate": 1.7762780887657576e-05, | |
| "loss": 0.7937, | |
| "mean_token_accuracy": 0.7642841154247401, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3254437869822485, | |
| "grad_norm": 0.37612205413583966, | |
| "learning_rate": 1.7071067811865477e-05, | |
| "loss": 0.7894, | |
| "mean_token_accuracy": 0.7649155785682786, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.35502958579881655, | |
| "grad_norm": 0.3690088505249562, | |
| "learning_rate": 1.6303906161279554e-05, | |
| "loss": 0.7999, | |
| "mean_token_accuracy": 0.761950455440128, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 0.3968668346625297, | |
| "learning_rate": 1.5469481581224274e-05, | |
| "loss": 0.7867, | |
| "mean_token_accuracy": 0.764712581061752, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.41420118343195267, | |
| "grad_norm": 0.37433783603713877, | |
| "learning_rate": 1.4576697415156818e-05, | |
| "loss": 0.7782, | |
| "mean_token_accuracy": 0.7675159747533142, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4437869822485207, | |
| "grad_norm": 0.33916682700714146, | |
| "learning_rate": 1.3635079705638298e-05, | |
| "loss": 0.777, | |
| "mean_token_accuracy": 0.7675317758209201, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.47337278106508873, | |
| "grad_norm": 0.34679361649836743, | |
| "learning_rate": 1.2654675551080724e-05, | |
| "loss": 0.7783, | |
| "mean_token_accuracy": 0.7665991894058249, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5029585798816568, | |
| "grad_norm": 0.3364148123003455, | |
| "learning_rate": 1.164594590280734e-05, | |
| "loss": 0.7785, | |
| "mean_token_accuracy": 0.7670094150622615, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.5325443786982249, | |
| "grad_norm": 0.3495734033672351, | |
| "learning_rate": 1.0619653946285948e-05, | |
| "loss": 0.7723, | |
| "mean_token_accuracy": 0.7682055741851116, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5621301775147929, | |
| "grad_norm": 0.3432574047378308, | |
| "learning_rate": 9.586750257511868e-06, | |
| "loss": 0.7669, | |
| "mean_token_accuracy": 0.7697794763971556, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.591715976331361, | |
| "grad_norm": 0.34082459115638813, | |
| "learning_rate": 8.558255959926533e-06, | |
| "loss": 0.7582, | |
| "mean_token_accuracy": 0.7714136401538525, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.591715976331361, | |
| "eval_loss": 0.7885270118713379, | |
| "eval_mean_token_accuracy": 0.7488056872787593, | |
| "eval_runtime": 2.2705, | |
| "eval_samples_per_second": 56.816, | |
| "eval_steps_per_second": 2.202, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.621301775147929, | |
| "grad_norm": 0.3500865265546408, | |
| "learning_rate": 7.545145128592009e-06, | |
| "loss": 0.759, | |
| "mean_token_accuracy": 0.7715606650120836, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.650887573964497, | |
| "grad_norm": 0.3153601426300555, | |
| "learning_rate": 6.558227696373617e-06, | |
| "loss": 0.7775, | |
| "mean_token_accuracy": 0.7664943003135225, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6804733727810651, | |
| "grad_norm": 0.3077751865894769, | |
| "learning_rate": 5.608034111526298e-06, | |
| "loss": 0.7636, | |
| "mean_token_accuracy": 0.7702647251741424, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.7100591715976331, | |
| "grad_norm": 0.2978423880943666, | |
| "learning_rate": 4.704702977392914e-06, | |
| "loss": 0.7511, | |
| "mean_token_accuracy": 0.7742831183944568, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.7396449704142012, | |
| "grad_norm": 0.3160810225440028, | |
| "learning_rate": 3.857872873103322e-06, | |
| "loss": 0.7594, | |
| "mean_token_accuracy": 0.7718046260565352, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 0.3193177717083944, | |
| "learning_rate": 3.0765795095517026e-06, | |
| "loss": 0.7623, | |
| "mean_token_accuracy": 0.7706338288914129, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.7988165680473372, | |
| "grad_norm": 0.301902197424481, | |
| "learning_rate": 2.369159318001937e-06, | |
| "loss": 0.7684, | |
| "mean_token_accuracy": 0.7688698616474955, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.8284023668639053, | |
| "grad_norm": 0.2941951800393329, | |
| "learning_rate": 1.743160500034443e-06, | |
| "loss": 0.7596, | |
| "mean_token_accuracy": 0.7716673849461525, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.8579881656804734, | |
| "grad_norm": 0.3117346710739024, | |
| "learning_rate": 1.2052624879351105e-06, | |
| "loss": 0.7595, | |
| "mean_token_accuracy": 0.7714754205949017, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.8875739644970414, | |
| "grad_norm": 0.2910628005200283, | |
| "learning_rate": 7.612046748871327e-07, | |
| "loss": 0.7505, | |
| "mean_token_accuracy": 0.7739860912523475, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.9171597633136095, | |
| "grad_norm": 0.28002315513328135, | |
| "learning_rate": 4.1572517541747294e-07, | |
| "loss": 0.741, | |
| "mean_token_accuracy": 0.7767315836804676, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.9467455621301775, | |
| "grad_norm": 0.2957619359463069, | |
| "learning_rate": 1.7251026952640583e-07, | |
| "loss": 0.754, | |
| "mean_token_accuracy": 0.7732596586427178, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.9763313609467456, | |
| "grad_norm": 0.2762985231070698, | |
| "learning_rate": 3.4155069933301535e-08, | |
| "loss": 0.7467, | |
| "mean_token_accuracy": 0.7749716424638167, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "mean_token_accuracy": 0.7704617250967111, | |
| "step": 169, | |
| "total_flos": 76973799899136.0, | |
| "train_loss": 0.8025016911636443, | |
| "train_runtime": 750.6152, | |
| "train_samples_per_second": 28.807, | |
| "train_steps_per_second": 0.225 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 169, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 76973799899136.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |