| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.00990099009901, | |
| "eval_steps": 7, | |
| "global_step": 76, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.039603960396039604, | |
| "eval_loss": 2.619887113571167, | |
| "eval_runtime": 1.2908, | |
| "eval_samples_per_second": 65.85, | |
| "eval_steps_per_second": 4.648, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1188118811881188, | |
| "grad_norm": 1.0429705381393433, | |
| "learning_rate": 3e-05, | |
| "loss": 2.4454, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2376237623762376, | |
| "grad_norm": 1.03085458278656, | |
| "learning_rate": 6e-05, | |
| "loss": 2.3957, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.27722772277227725, | |
| "eval_loss": 2.1992604732513428, | |
| "eval_runtime": 1.2817, | |
| "eval_samples_per_second": 66.32, | |
| "eval_steps_per_second": 4.681, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3564356435643564, | |
| "grad_norm": 0.7208812236785889, | |
| "learning_rate": 9e-05, | |
| "loss": 2.0654, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.4752475247524752, | |
| "grad_norm": 0.8885481357574463, | |
| "learning_rate": 9.977359612865423e-05, | |
| "loss": 2.0394, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5544554455445545, | |
| "eval_loss": 1.8464581966400146, | |
| "eval_runtime": 1.2853, | |
| "eval_samples_per_second": 66.133, | |
| "eval_steps_per_second": 4.668, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.594059405940594, | |
| "grad_norm": 0.7464063763618469, | |
| "learning_rate": 9.859057841617709e-05, | |
| "loss": 1.8413, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.7128712871287128, | |
| "grad_norm": 0.7601683139801025, | |
| "learning_rate": 9.641839665080363e-05, | |
| "loss": 1.6764, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.8316831683168316, | |
| "grad_norm": 0.8399017453193665, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 1.6186, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8316831683168316, | |
| "eval_loss": 1.5880630016326904, | |
| "eval_runtime": 1.2879, | |
| "eval_samples_per_second": 65.997, | |
| "eval_steps_per_second": 4.659, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.9504950495049505, | |
| "grad_norm": 0.7568883299827576, | |
| "learning_rate": 8.930265473713938e-05, | |
| "loss": 1.5255, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.0693069306930694, | |
| "grad_norm": 0.6296226978302002, | |
| "learning_rate": 8.450395057410561e-05, | |
| "loss": 1.6128, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.108910891089109, | |
| "eval_loss": 1.4710044860839844, | |
| "eval_runtime": 1.2852, | |
| "eval_samples_per_second": 66.136, | |
| "eval_steps_per_second": 4.668, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.188118811881188, | |
| "grad_norm": 0.7296110391616821, | |
| "learning_rate": 7.900284547855991e-05, | |
| "loss": 1.3108, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.306930693069307, | |
| "grad_norm": 0.6805970072746277, | |
| "learning_rate": 7.291132608637052e-05, | |
| "loss": 1.2599, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.386138613861386, | |
| "eval_loss": 1.4217455387115479, | |
| "eval_runtime": 1.2877, | |
| "eval_samples_per_second": 66.008, | |
| "eval_steps_per_second": 4.659, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4257425742574257, | |
| "grad_norm": 0.7158127427101135, | |
| "learning_rate": 6.635339816587109e-05, | |
| "loss": 1.2425, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.5445544554455446, | |
| "grad_norm": 0.7889519929885864, | |
| "learning_rate": 5.946256221802051e-05, | |
| "loss": 1.2471, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.6633663366336635, | |
| "grad_norm": 1.0730153322219849, | |
| "learning_rate": 5.2379095791187124e-05, | |
| "loss": 1.1683, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6633663366336635, | |
| "eval_loss": 1.3953288793563843, | |
| "eval_runtime": 1.2886, | |
| "eval_samples_per_second": 65.965, | |
| "eval_steps_per_second": 4.656, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.7821782178217822, | |
| "grad_norm": 0.8776001930236816, | |
| "learning_rate": 4.5247197834790876e-05, | |
| "loss": 1.1555, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.900990099009901, | |
| "grad_norm": 0.8503000140190125, | |
| "learning_rate": 3.821205322452863e-05, | |
| "loss": 1.2638, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9405940594059405, | |
| "eval_loss": 1.384788155555725, | |
| "eval_runtime": 1.2856, | |
| "eval_samples_per_second": 66.119, | |
| "eval_steps_per_second": 4.667, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.01980198019802, | |
| "grad_norm": 1.8872629404067993, | |
| "learning_rate": 3.141687721698363e-05, | |
| "loss": 1.4865, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.1386138613861387, | |
| "grad_norm": 0.8172866106033325, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.9957, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.217821782178218, | |
| "eval_loss": 1.3781214952468872, | |
| "eval_runtime": 1.286, | |
| "eval_samples_per_second": 66.094, | |
| "eval_steps_per_second": 4.665, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.2574257425742577, | |
| "grad_norm": 0.8602892160415649, | |
| "learning_rate": 1.9092050688969738e-05, | |
| "loss": 1.0049, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.376237623762376, | |
| "grad_norm": 0.8723423480987549, | |
| "learning_rate": 1.3813298094746491e-05, | |
| "loss": 1.0431, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.495049504950495, | |
| "grad_norm": 0.8783928155899048, | |
| "learning_rate": 9.271202397483215e-06, | |
| "loss": 0.9437, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.495049504950495, | |
| "eval_loss": 1.3786593675613403, | |
| "eval_runtime": 1.3013, | |
| "eval_samples_per_second": 65.322, | |
| "eval_steps_per_second": 4.611, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.613861386138614, | |
| "grad_norm": 0.9128199815750122, | |
| "learning_rate": 5.558227567253832e-06, | |
| "loss": 1.023, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.7326732673267324, | |
| "grad_norm": 0.9300925731658936, | |
| "learning_rate": 2.7499590642665774e-06, | |
| "loss": 1.0016, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.772277227722772, | |
| "eval_loss": 1.3773143291473389, | |
| "eval_runtime": 1.2839, | |
| "eval_samples_per_second": 66.203, | |
| "eval_steps_per_second": 4.673, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.8514851485148514, | |
| "grad_norm": 0.9618849754333496, | |
| "learning_rate": 9.035651368646648e-07, | |
| "loss": 1.0099, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.9702970297029703, | |
| "grad_norm": 0.9259308576583862, | |
| "learning_rate": 5.663304084960186e-08, | |
| "loss": 0.9467, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 76, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 7, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.1612707507181978e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |