| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 280, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03571428571428571, | |
| "grad_norm": 19.5, | |
| "learning_rate": 1.5838110992819075e-05, | |
| "loss": 4.4082, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07142857142857142, | |
| "grad_norm": 11.6875, | |
| "learning_rate": 3.5635749733842914e-05, | |
| "loss": 3.2859, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10714285714285714, | |
| "grad_norm": 8.1875, | |
| "learning_rate": 5.5433388474866766e-05, | |
| "loss": 2.8128, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": 6.3125, | |
| "learning_rate": 7.52310272158906e-05, | |
| "loss": 2.6378, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.17857142857142858, | |
| "grad_norm": 5.3125, | |
| "learning_rate": 9.502866595691445e-05, | |
| "loss": 2.4924, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.21428571428571427, | |
| "grad_norm": 5.84375, | |
| "learning_rate": 0.0001148263046979383, | |
| "loss": 2.4994, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 5.125, | |
| "learning_rate": 0.00013462394343896213, | |
| "loss": 2.4877, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 5.3125, | |
| "learning_rate": 0.00013855579078350677, | |
| "loss": 2.5123, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.32142857142857145, | |
| "grad_norm": 4.03125, | |
| "learning_rate": 0.0001384433896787117, | |
| "loss": 2.453, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.35714285714285715, | |
| "grad_norm": 4.5625, | |
| "learning_rate": 0.00013824472478409896, | |
| "loss": 2.4429, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.39285714285714285, | |
| "grad_norm": 3.75, | |
| "learning_rate": 0.00013796012675777958, | |
| "loss": 2.4012, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": 3.234375, | |
| "learning_rate": 0.00013759006928508162, | |
| "loss": 2.351, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4642857142857143, | |
| "grad_norm": 3.71875, | |
| "learning_rate": 0.00013713516829014772, | |
| "loss": 2.3272, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 3.875, | |
| "learning_rate": 0.00013659618091079015, | |
| "loss": 2.3015, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5357142857142857, | |
| "grad_norm": 3.171875, | |
| "learning_rate": 0.00013597400423830962, | |
| "loss": 2.2773, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 3.359375, | |
| "learning_rate": 0.00013526967382437543, | |
| "loss": 2.2415, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6071428571428571, | |
| "grad_norm": 2.9375, | |
| "learning_rate": 0.00013448436195745173, | |
| "loss": 2.1806, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.6428571428571429, | |
| "grad_norm": 2.8125, | |
| "learning_rate": 0.00013361937571163926, | |
| "loss": 2.1731, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6785714285714286, | |
| "grad_norm": 3.015625, | |
| "learning_rate": 0.00013267615477117944, | |
| "loss": 2.2264, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 2.75, | |
| "learning_rate": 0.0001316562690342419, | |
| "loss": 2.1964, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 2.71875, | |
| "learning_rate": 0.00013056141599998415, | |
| "loss": 2.2052, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.7857142857142857, | |
| "grad_norm": 3.21875, | |
| "learning_rate": 0.00012939341794323162, | |
| "loss": 2.1698, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.8214285714285714, | |
| "grad_norm": 2.734375, | |
| "learning_rate": 0.00012815421888148093, | |
| "loss": 2.1723, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 2.515625, | |
| "learning_rate": 0.0001268458813392749, | |
| "loss": 2.1478, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8928571428571429, | |
| "grad_norm": 2.546875, | |
| "learning_rate": 0.00012547058291533382, | |
| "loss": 2.0849, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.9285714285714286, | |
| "grad_norm": 2.640625, | |
| "learning_rate": 0.00012403061265815721, | |
| "loss": 2.1297, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.9642857142857143, | |
| "grad_norm": 2.5, | |
| "learning_rate": 0.00012252836725612872, | |
| "loss": 2.0767, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.34375, | |
| "learning_rate": 0.00012096634704846468, | |
| "loss": 2.1226, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.0443668365478516, | |
| "eval_runtime": 0.4691, | |
| "eval_samples_per_second": 10.658, | |
| "eval_steps_per_second": 10.658, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.0357142857142858, | |
| "grad_norm": 2.5625, | |
| "learning_rate": 0.00011934715186364626, | |
| "loss": 1.7431, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.0714285714285714, | |
| "grad_norm": 2.390625, | |
| "learning_rate": 0.00011767347669226159, | |
| "loss": 1.6931, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1071428571428572, | |
| "grad_norm": 2.40625, | |
| "learning_rate": 0.00011594810720145982, | |
| "loss": 1.6984, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 2.46875, | |
| "learning_rate": 0.00011417391509848311, | |
| "loss": 1.7018, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.1785714285714286, | |
| "grad_norm": 2.140625, | |
| "learning_rate": 0.00011235385335099333, | |
| "loss": 1.7014, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.2142857142857142, | |
| "grad_norm": 2.65625, | |
| "learning_rate": 0.00011049095127214879, | |
| "loss": 1.7539, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.65625, | |
| "learning_rate": 0.00010858830947861153, | |
| "loss": 1.7273, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.2857142857142856, | |
| "grad_norm": 2.375, | |
| "learning_rate": 0.00010664909472987668, | |
| "loss": 1.7267, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.3214285714285714, | |
| "grad_norm": 2.484375, | |
| "learning_rate": 0.00010467653465751409, | |
| "loss": 1.7098, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.3571428571428572, | |
| "grad_norm": 2.359375, | |
| "learning_rate": 0.00010267391239309385, | |
| "loss": 1.7123, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.3928571428571428, | |
| "grad_norm": 2.25, | |
| "learning_rate": 0.00010064456110373801, | |
| "loss": 1.7539, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 2.34375, | |
| "learning_rate": 9.859185844439268e-05, | |
| "loss": 1.7102, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.4642857142857144, | |
| "grad_norm": 2.203125, | |
| "learning_rate": 9.651922093605485e-05, | |
| "loss": 1.6981, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 2.171875, | |
| "learning_rate": 9.443009827931028e-05, | |
| "loss": 1.7112, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.5357142857142856, | |
| "grad_norm": 2.234375, | |
| "learning_rate": 9.232796761264735e-05, | |
| "loss": 1.7439, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.5714285714285714, | |
| "grad_norm": 2.21875, | |
| "learning_rate": 9.021632772510318e-05, | |
| "loss": 1.7149, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.6071428571428572, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 8.809869323287461e-05, | |
| "loss": 1.6947, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.6428571428571428, | |
| "grad_norm": 2.296875, | |
| "learning_rate": 8.597858872958643e-05, | |
| "loss": 1.6722, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.6785714285714286, | |
| "grad_norm": 2.140625, | |
| "learning_rate": 8.385954291995294e-05, | |
| "loss": 1.6419, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 2.0, | |
| "learning_rate": 8.174508274659743e-05, | |
| "loss": 1.6508, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 2.03125, | |
| "learning_rate": 7.963872751980418e-05, | |
| "loss": 1.6213, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.7857142857142856, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 7.754398305997372e-05, | |
| "loss": 1.6257, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.8214285714285714, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 7.546433586253086e-05, | |
| "loss": 1.5915, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.8571428571428572, | |
| "grad_norm": 1.890625, | |
| "learning_rate": 7.340324729499708e-05, | |
| "loss": 1.6334, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.8928571428571428, | |
| "grad_norm": 1.96875, | |
| "learning_rate": 7.136414783588594e-05, | |
| "loss": 1.6345, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.9285714285714286, | |
| "grad_norm": 2.015625, | |
| "learning_rate": 6.935043136501011e-05, | |
| "loss": 1.6109, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.9642857142857144, | |
| "grad_norm": 2.09375, | |
| "learning_rate": 6.736544951470336e-05, | |
| "loss": 1.6274, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 6.541250609135935e-05, | |
| "loss": 1.6177, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.8396857976913452, | |
| "eval_runtime": 0.3556, | |
| "eval_samples_per_second": 14.061, | |
| "eval_steps_per_second": 14.061, | |
| "step": 280 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 420, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.6551816128692224e+17, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |