| { | |
| "best_metric": 2.0174262523651123, | |
| "best_model_checkpoint": "./qlora-out-kimiko-test2/checkpoint-80", | |
| "epoch": 2.0678513731825525, | |
| "eval_steps": 10, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1e-05, | |
| "loss": 1.5707, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2e-05, | |
| "loss": 1.5621, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4812, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4e-05, | |
| "loss": 1.5197, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5567, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4645, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6122, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5596, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5608, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5456, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_loss": 2.098437547683716, | |
| "eval_runtime": 119.6161, | |
| "eval_samples_per_second": 1.555, | |
| "eval_steps_per_second": 0.777, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5645, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 5e-05, | |
| "loss": 1.538, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6388, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4943, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5469, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6149, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5345, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4903, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5499, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5934, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_loss": 2.066981077194214, | |
| "eval_runtime": 119.5781, | |
| "eval_samples_per_second": 1.555, | |
| "eval_steps_per_second": 0.778, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4554, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5512, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4636, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5398, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5623, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4658, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4723, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 5e-05, | |
| "loss": 1.432, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4814, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4924, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_loss": 2.046339750289917, | |
| "eval_runtime": 119.5771, | |
| "eval_samples_per_second": 1.555, | |
| "eval_steps_per_second": 0.778, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5809, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4803, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4878, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3871, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5151, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4212, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6284, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5002, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4452, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4399, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "eval_loss": 2.0354697704315186, | |
| "eval_runtime": 119.5875, | |
| "eval_samples_per_second": 1.555, | |
| "eval_steps_per_second": 0.778, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 5e-05, | |
| "loss": 1.465, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4199, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5403, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4499, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5751, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4809, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5022, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4663, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4435, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4246, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 2.0276732444763184, | |
| "eval_runtime": 119.5811, | |
| "eval_samples_per_second": 1.555, | |
| "eval_steps_per_second": 0.778, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4877, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4066, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3559, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5591, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4942, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4685, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4165, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3995, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3931, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4234, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_loss": 2.0228564739227295, | |
| "eval_runtime": 119.5715, | |
| "eval_samples_per_second": 1.556, | |
| "eval_steps_per_second": 0.778, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4732, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4349, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4548, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 5e-05, | |
| "loss": 1.48, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3789, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3915, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3789, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5206, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4851, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5251, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_loss": 2.0199856758117676, | |
| "eval_runtime": 119.5994, | |
| "eval_samples_per_second": 1.555, | |
| "eval_steps_per_second": 0.778, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4152, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4262, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5563, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 5e-05, | |
| "loss": 1.42, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4407, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3953, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4186, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3863, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3907, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5147, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_loss": 2.0174262523651123, | |
| "eval_runtime": 119.6245, | |
| "eval_samples_per_second": 1.555, | |
| "eval_steps_per_second": 0.777, | |
| "step": 80 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 114, | |
| "num_train_epochs": 3, | |
| "save_steps": 10, | |
| "total_flos": 3.667626583051469e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |