| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.72972972972973, | |
| "eval_steps": 500, | |
| "global_step": 180, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05405405405405406, | |
| "grad_norm": 2.03125, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 3.0425, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.2702702702702703, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 5.555555555555556e-05, | |
| "loss": 3.0524, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "grad_norm": 2.0625, | |
| "learning_rate": 0.00011111111111111112, | |
| "loss": 2.893, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.8108108108108109, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 2.5077, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.972972972972973, | |
| "eval_loss": 2.778700351715088, | |
| "eval_runtime": 0.5506, | |
| "eval_samples_per_second": 18.162, | |
| "eval_steps_per_second": 1.816, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 2.546875, | |
| "learning_rate": 0.00019992479525042303, | |
| "loss": 2.2381, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 0.00019908004033648453, | |
| "loss": 1.9825, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "grad_norm": 82.0, | |
| "learning_rate": 0.00019730448705798239, | |
| "loss": 1.8071, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.8918918918918919, | |
| "grad_norm": 0.72265625, | |
| "learning_rate": 0.00019461481568757506, | |
| "loss": 1.6701, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.600048065185547, | |
| "eval_runtime": 0.5541, | |
| "eval_samples_per_second": 18.046, | |
| "eval_steps_per_second": 1.805, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 0.5078125, | |
| "learning_rate": 0.0001910362940966147, | |
| "loss": 1.5495, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.4324324324324325, | |
| "grad_norm": 0.439453125, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 1.4905, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "grad_norm": 0.376953125, | |
| "learning_rate": 0.00018135520702629675, | |
| "loss": 1.4373, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.972972972972973, | |
| "grad_norm": 0.25, | |
| "learning_rate": 0.00017534358963276607, | |
| "loss": 1.3757, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.972972972972973, | |
| "eval_loss": 2.521566152572632, | |
| "eval_runtime": 0.5468, | |
| "eval_samples_per_second": 18.289, | |
| "eval_steps_per_second": 1.829, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 3.2432432432432434, | |
| "grad_norm": 0.474609375, | |
| "learning_rate": 0.0001686241637868734, | |
| "loss": 1.3452, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.5135135135135136, | |
| "grad_norm": 0.6171875, | |
| "learning_rate": 0.0001612600545193203, | |
| "loss": 1.315, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 3.7837837837837838, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 0.00015332044328016914, | |
| "loss": 1.2905, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.51369309425354, | |
| "eval_runtime": 0.5468, | |
| "eval_samples_per_second": 18.287, | |
| "eval_steps_per_second": 1.829, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 4.054054054054054, | |
| "grad_norm": 0.345703125, | |
| "learning_rate": 0.00014487991802004623, | |
| "loss": 1.2801, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 4.324324324324325, | |
| "grad_norm": 0.482421875, | |
| "learning_rate": 0.00013601777248047105, | |
| "loss": 1.2552, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.594594594594595, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 0.00012681726127606376, | |
| "loss": 1.2395, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 4.864864864864865, | |
| "grad_norm": 0.45703125, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 1.2291, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.972972972972973, | |
| "eval_loss": 2.5113296508789062, | |
| "eval_runtime": 0.5537, | |
| "eval_samples_per_second": 18.061, | |
| "eval_steps_per_second": 1.806, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.135135135135135, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 0.0001077492420671931, | |
| "loss": 1.2166, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 5.405405405405405, | |
| "grad_norm": 0.431640625, | |
| "learning_rate": 9.806086682281758e-05, | |
| "loss": 1.1963, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.675675675675675, | |
| "grad_norm": 0.408203125, | |
| "learning_rate": 8.839070858747697e-05, | |
| "loss": 1.2011, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 5.945945945945946, | |
| "grad_norm": 0.390625, | |
| "learning_rate": 7.882961277705895e-05, | |
| "loss": 1.1946, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.5234570503234863, | |
| "eval_runtime": 0.5594, | |
| "eval_samples_per_second": 17.877, | |
| "eval_steps_per_second": 1.788, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 6.216216216216216, | |
| "grad_norm": 0.4765625, | |
| "learning_rate": 6.94674002304887e-05, | |
| "loss": 1.1733, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 6.486486486486487, | |
| "grad_norm": 0.3125, | |
| "learning_rate": 6.039202339608432e-05, | |
| "loss": 1.1803, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.756756756756757, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 5.168874007033615e-05, | |
| "loss": 1.1618, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 6.972972972972973, | |
| "eval_loss": 2.5299763679504395, | |
| "eval_runtime": 0.5604, | |
| "eval_samples_per_second": 17.843, | |
| "eval_steps_per_second": 1.784, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 7.027027027027027, | |
| "grad_norm": 0.36328125, | |
| "learning_rate": 4.343931245134616e-05, | |
| "loss": 1.1584, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 7.297297297297297, | |
| "grad_norm": 0.357421875, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 1.1484, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 7.5675675675675675, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 2.8607026544210114e-05, | |
| "loss": 1.1477, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.837837837837838, | |
| "grad_norm": 0.359375, | |
| "learning_rate": 2.2163508807583998e-05, | |
| "loss": 1.1521, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.5334644317626953, | |
| "eval_runtime": 0.549, | |
| "eval_samples_per_second": 18.214, | |
| "eval_steps_per_second": 1.821, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 8.108108108108109, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 1.6451218858706374e-05, | |
| "loss": 1.1642, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.378378378378379, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 1.1523820282334219e-05, | |
| "loss": 1.1391, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 8.64864864864865, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 7.427603073110967e-06, | |
| "loss": 1.1491, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.91891891891892, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 4.20104876845111e-06, | |
| "loss": 1.147, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 8.972972972972974, | |
| "eval_loss": 2.534287929534912, | |
| "eval_runtime": 0.5609, | |
| "eval_samples_per_second": 17.829, | |
| "eval_steps_per_second": 1.783, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 9.18918918918919, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 1.874468937261531e-06, | |
| "loss": 1.1527, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 9.45945945945946, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 4.6972042068341714e-07, | |
| "loss": 1.1486, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 9.72972972972973, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 0.0, | |
| "loss": 1.14, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.72972972972973, | |
| "eval_loss": 2.534336805343628, | |
| "eval_runtime": 0.5494, | |
| "eval_samples_per_second": 18.202, | |
| "eval_steps_per_second": 1.82, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.72972972972973, | |
| "step": 180, | |
| "total_flos": 1.0572720102178816e+17, | |
| "train_loss": 1.4424436648686727, | |
| "train_runtime": 971.2544, | |
| "train_samples_per_second": 9.019, | |
| "train_steps_per_second": 0.185 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0572720102178816e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |