| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0654205607476634, | |
| "eval_steps": 25, | |
| "global_step": 41, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07476635514018691, | |
| "grad_norm": 130.8704376220703, | |
| "learning_rate": 5e-05, | |
| "loss": 122.9282, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07476635514018691, | |
| "eval_loss": 7.675048828125, | |
| "eval_runtime": 4.5782, | |
| "eval_samples_per_second": 5.024, | |
| "eval_steps_per_second": 2.621, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.14953271028037382, | |
| "grad_norm": 130.634521484375, | |
| "learning_rate": 0.0001, | |
| "loss": 122.9184, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.22429906542056074, | |
| "grad_norm": 130.010009765625, | |
| "learning_rate": 9.983786540671051e-05, | |
| "loss": 124.9848, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.29906542056074764, | |
| "grad_norm": 186.89804077148438, | |
| "learning_rate": 9.935251313189564e-05, | |
| "loss": 92.2971, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.37383177570093457, | |
| "grad_norm": 191.04327392578125, | |
| "learning_rate": 9.85470908713026e-05, | |
| "loss": 61.2594, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.4485981308411215, | |
| "grad_norm": 173.75599670410156, | |
| "learning_rate": 9.742682209735727e-05, | |
| "loss": 34.2174, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.5233644859813084, | |
| "grad_norm": 87.42670440673828, | |
| "learning_rate": 9.599897218294122e-05, | |
| "loss": 16.7545, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.5981308411214953, | |
| "grad_norm": 33.45049285888672, | |
| "learning_rate": 9.42728012826605e-05, | |
| "loss": 12.7406, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.6728971962616822, | |
| "grad_norm": 64.37686157226562, | |
| "learning_rate": 9.225950427718975e-05, | |
| "loss": 12.6661, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.7476635514018691, | |
| "grad_norm": 24.238252639770508, | |
| "learning_rate": 8.997213817017507e-05, | |
| "loss": 11.2995, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.822429906542056, | |
| "grad_norm": 20.897985458374023, | |
| "learning_rate": 8.742553740855506e-05, | |
| "loss": 11.7569, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.897196261682243, | |
| "grad_norm": 13.548663139343262, | |
| "learning_rate": 8.463621767547998e-05, | |
| "loss": 10.9101, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.9719626168224299, | |
| "grad_norm": 29.94806480407715, | |
| "learning_rate": 8.162226877976887e-05, | |
| "loss": 12.1377, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.0467289719626167, | |
| "grad_norm": 14.382259368896484, | |
| "learning_rate": 7.840323733655778e-05, | |
| "loss": 11.0721, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.1214953271028036, | |
| "grad_norm": 17.59343147277832, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 11.1833, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.1962616822429906, | |
| "grad_norm": 16.972679138183594, | |
| "learning_rate": 7.143462807015271e-05, | |
| "loss": 11.4135, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.2710280373831775, | |
| "grad_norm": 22.415023803710938, | |
| "learning_rate": 6.773024435212678e-05, | |
| "loss": 10.9927, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.3457943925233644, | |
| "grad_norm": 8.642698287963867, | |
| "learning_rate": 6.391087319582264e-05, | |
| "loss": 10.8987, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.4205607476635513, | |
| "grad_norm": 26.28803825378418, | |
| "learning_rate": 6.0001284688802226e-05, | |
| "loss": 11.544, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.4953271028037383, | |
| "grad_norm": 27.87239646911621, | |
| "learning_rate": 5.602683401276615e-05, | |
| "loss": 11.4118, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.5700934579439252, | |
| "grad_norm": 12.727015495300293, | |
| "learning_rate": 5.201329700547076e-05, | |
| "loss": 10.6913, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.644859813084112, | |
| "grad_norm": 15.245328903198242, | |
| "learning_rate": 4.798670299452926e-05, | |
| "loss": 11.0664, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.719626168224299, | |
| "grad_norm": 27.16010284423828, | |
| "learning_rate": 4.397316598723385e-05, | |
| "loss": 12.1439, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.794392523364486, | |
| "grad_norm": 26.858306884765625, | |
| "learning_rate": 3.9998715311197785e-05, | |
| "loss": 11.9096, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.8691588785046729, | |
| "grad_norm": 19.373350143432617, | |
| "learning_rate": 3.608912680417737e-05, | |
| "loss": 11.1433, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.8691588785046729, | |
| "eval_loss": 0.7009603977203369, | |
| "eval_runtime": 4.6575, | |
| "eval_samples_per_second": 4.938, | |
| "eval_steps_per_second": 2.576, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.9439252336448598, | |
| "grad_norm": 9.355457305908203, | |
| "learning_rate": 3.226975564787322e-05, | |
| "loss": 11.0052, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.0186915887850465, | |
| "grad_norm": 13.675053596496582, | |
| "learning_rate": 2.8565371929847284e-05, | |
| "loss": 11.0252, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.0934579439252334, | |
| "grad_norm": 14.849433898925781, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 10.9197, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.1682242990654204, | |
| "grad_norm": 12.167821884155273, | |
| "learning_rate": 2.1596762663442218e-05, | |
| "loss": 10.4549, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.2429906542056073, | |
| "grad_norm": 14.887701988220215, | |
| "learning_rate": 1.837773122023114e-05, | |
| "loss": 11.138, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.317757009345794, | |
| "grad_norm": 13.879959106445312, | |
| "learning_rate": 1.536378232452003e-05, | |
| "loss": 11.0227, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.392523364485981, | |
| "grad_norm": 10.067033767700195, | |
| "learning_rate": 1.257446259144494e-05, | |
| "loss": 10.97, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.467289719626168, | |
| "grad_norm": 13.370949745178223, | |
| "learning_rate": 1.0027861829824952e-05, | |
| "loss": 10.5925, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.542056074766355, | |
| "grad_norm": 9.568769454956055, | |
| "learning_rate": 7.740495722810271e-06, | |
| "loss": 11.3019, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.616822429906542, | |
| "grad_norm": 13.338472366333008, | |
| "learning_rate": 5.727198717339511e-06, | |
| "loss": 11.2045, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.691588785046729, | |
| "grad_norm": 17.480472564697266, | |
| "learning_rate": 4.001027817058789e-06, | |
| "loss": 10.5578, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.7663551401869158, | |
| "grad_norm": 9.267881393432617, | |
| "learning_rate": 2.573177902642726e-06, | |
| "loss": 10.4758, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.8411214953271027, | |
| "grad_norm": 14.049615859985352, | |
| "learning_rate": 1.4529091286973995e-06, | |
| "loss": 11.4235, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.9158878504672896, | |
| "grad_norm": 17.158531188964844, | |
| "learning_rate": 6.474868681043578e-07, | |
| "loss": 11.0862, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.9906542056074765, | |
| "grad_norm": 11.268014907836914, | |
| "learning_rate": 1.6213459328950352e-07, | |
| "loss": 10.8354, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.0654205607476634, | |
| "grad_norm": 6.079852104187012, | |
| "learning_rate": 0.0, | |
| "loss": 10.5835, | |
| "step": 41 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 41, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0756512736817971e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |