| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 90, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0, | |
| "loss": 1.842, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00012618595071429146, | |
| "loss": 1.3311, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4462, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2415, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1317, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002, | |
| "loss": 1.039, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9913, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8591, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.896, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.01621675491333, | |
| "eval_runtime": 5.155, | |
| "eval_samples_per_second": 9.699, | |
| "eval_steps_per_second": 0.388, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.686, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5042, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4833, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5426, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4841, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4638, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3123, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2776, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2277, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1135010719299316, | |
| "eval_runtime": 5.2455, | |
| "eval_samples_per_second": 9.532, | |
| "eval_steps_per_second": 0.381, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1669, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1097, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1085, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1307, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0898, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.056, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0965, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0987, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0717, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.1909135580062866, | |
| "eval_runtime": 5.2212, | |
| "eval_samples_per_second": 9.576, | |
| "eval_steps_per_second": 0.383, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0384, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0343, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.04, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0236, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0336, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0324, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0326, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0368, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0271, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.4077610969543457, | |
| "eval_runtime": 5.2331, | |
| "eval_samples_per_second": 9.555, | |
| "eval_steps_per_second": 0.382, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0099, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0154, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0141, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0121, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0075, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0167, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0204, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0096, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0083, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.4836626052856445, | |
| "eval_runtime": 5.2599, | |
| "eval_samples_per_second": 9.506, | |
| "eval_steps_per_second": 0.38, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.005, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0089, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0051, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0073, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0038, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.007, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0044, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0023, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0087, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.5063480138778687, | |
| "eval_runtime": 5.1771, | |
| "eval_samples_per_second": 9.658, | |
| "eval_steps_per_second": 0.386, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0021, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0024, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0087, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0046, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0045, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0024, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0096, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0058, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0087, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.4928628206253052, | |
| "eval_runtime": 5.1608, | |
| "eval_samples_per_second": 9.688, | |
| "eval_steps_per_second": 0.388, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0023, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0025, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.003, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 7.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0043, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0071, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0039, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0092, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0039, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0011, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.538659930229187, | |
| "eval_runtime": 5.1718, | |
| "eval_samples_per_second": 9.668, | |
| "eval_steps_per_second": 0.387, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0021, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.006, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0026, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0029, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0058, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 8.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0009, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0025, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0061, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0008, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.5352141857147217, | |
| "eval_runtime": 5.2355, | |
| "eval_samples_per_second": 9.55, | |
| "eval_steps_per_second": 0.382, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0021, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0028, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0087, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0069, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0052, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0021, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0031, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 0.0002, | |
| "loss": 0.011, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0032, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.4710699319839478, | |
| "eval_runtime": 5.2002, | |
| "eval_samples_per_second": 9.615, | |
| "eval_steps_per_second": 0.385, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 90, | |
| "total_flos": 1.6878235281679974e+17, | |
| "train_loss": 0.18127101581647165, | |
| "train_runtime": 635.4362, | |
| "train_samples_per_second": 4.454, | |
| "train_steps_per_second": 0.142 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 90, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 50000, | |
| "total_flos": 1.6878235281679974e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |