| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.857142857142857, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 4.689024448394775, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.1081, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 5.0324506759643555, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.1116, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 4.365543842315674, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1051, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 2.0849297046661377, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 1.0214, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 4.037038803100586, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 1.0044, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 3.972327947616577, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 1.0263, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 3.0667643547058105, | |
| "learning_rate": 9.468163201617063e-06, | |
| "loss": 1.0135, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 2.5635499954223633, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 0.9599, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 2.112760305404663, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 0.9647, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 1.69428288936615, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 0.9203, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0476190476190477, | |
| "grad_norm": 1.9577350616455078, | |
| "learning_rate": 7.985792958513932e-06, | |
| "loss": 1.3375, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 1.1105504035949707, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.866, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2380952380952381, | |
| "grad_norm": 1.216810703277588, | |
| "learning_rate": 6.980398830195785e-06, | |
| "loss": 0.8611, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 1.2832354307174683, | |
| "learning_rate": 6.434016163555452e-06, | |
| "loss": 0.8567, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.8837525248527527, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.7996, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 0.8987547755241394, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 0.7913, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.619047619047619, | |
| "grad_norm": 1.1900713443756104, | |
| "learning_rate": 4.7092758554476215e-06, | |
| "loss": 0.8828, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.9665733575820923, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.8782, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8095238095238095, | |
| "grad_norm": 0.6934548616409302, | |
| "learning_rate": 3.5659838364445505e-06, | |
| "loss": 0.7534, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.8050403594970703, | |
| "learning_rate": 3.019601169804216e-06, | |
| "loss": 0.8343, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.1305651664733887, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 1.1694, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.7208789587020874, | |
| "learning_rate": 2.0142070414860704e-06, | |
| "loss": 0.7511, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 0.6582717895507812, | |
| "learning_rate": 1.5687918106563326e-06, | |
| "loss": 0.7899, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.7116344571113586, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.771, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.5961452126502991, | |
| "learning_rate": 8.225609429353187e-07, | |
| "loss": 0.7756, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 0.6315088868141174, | |
| "learning_rate": 5.318367983829393e-07, | |
| "loss": 0.7748, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.5613293647766113, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.7243, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.5505610108375549, | |
| "learning_rate": 1.3477564710088097e-07, | |
| "loss": 0.7717, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 0.5840697288513184, | |
| "learning_rate": 3.3808211290284886e-08, | |
| "loss": 0.7599, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.5845107436180115, | |
| "learning_rate": 0.0, | |
| "loss": 0.7456, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "step": 30, | |
| "total_flos": 17389270925312.0, | |
| "train_loss": 0.9043172419071197, | |
| "train_runtime": 792.0568, | |
| "train_samples_per_second": 3.788, | |
| "train_steps_per_second": 0.038 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 17389270925312.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |