| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.556149732620321, | |
| "eval_steps": 400, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0213903743315508, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0, | |
| "loss": 1.9241, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.21390374331550802, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0, | |
| "loss": 1.7796, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.42780748663101603, | |
| "grad_norm": 43.9833984375, | |
| "learning_rate": 4e-08, | |
| "loss": 1.8211, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6417112299465241, | |
| "grad_norm": 40.75557327270508, | |
| "learning_rate": 4.4e-07, | |
| "loss": 1.733, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8556149732620321, | |
| "grad_norm": 33.69241714477539, | |
| "learning_rate": 8.400000000000001e-07, | |
| "loss": 1.4004, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0695187165775402, | |
| "grad_norm": 32.86851501464844, | |
| "learning_rate": 1.2400000000000002e-06, | |
| "loss": 1.3416, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.2834224598930482, | |
| "grad_norm": 27.803329467773438, | |
| "learning_rate": 1.6400000000000002e-06, | |
| "loss": 1.081, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.4973262032085561, | |
| "grad_norm": 31.654834747314453, | |
| "learning_rate": 2.04e-06, | |
| "loss": 1.05, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.7112299465240641, | |
| "grad_norm": 29.529935836791992, | |
| "learning_rate": 2.4400000000000004e-06, | |
| "loss": 0.9944, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.9251336898395723, | |
| "grad_norm": 29.650171279907227, | |
| "learning_rate": 2.84e-06, | |
| "loss": 0.8562, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.1390374331550803, | |
| "grad_norm": 19.60784912109375, | |
| "learning_rate": 3.2400000000000003e-06, | |
| "loss": 0.5071, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 40.14349365234375, | |
| "learning_rate": 3.6400000000000003e-06, | |
| "loss": 0.2566, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.5668449197860963, | |
| "grad_norm": 38.05447006225586, | |
| "learning_rate": 4.04e-06, | |
| "loss": 0.2548, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.7807486631016043, | |
| "grad_norm": 48.719520568847656, | |
| "learning_rate": 4.440000000000001e-06, | |
| "loss": 0.2113, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.9946524064171123, | |
| "grad_norm": 82.81842803955078, | |
| "learning_rate": 4.84e-06, | |
| "loss": 0.3946, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.2085561497326203, | |
| "grad_norm": 20.48181915283203, | |
| "learning_rate": 5.240000000000001e-06, | |
| "loss": 0.09, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.4224598930481283, | |
| "grad_norm": 5.467124938964844, | |
| "learning_rate": 5.64e-06, | |
| "loss": 0.07, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 27.81447410583496, | |
| "learning_rate": 6.040000000000001e-06, | |
| "loss": 0.0837, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.8502673796791442, | |
| "grad_norm": 23.266380310058594, | |
| "learning_rate": 6.440000000000001e-06, | |
| "loss": 0.1036, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.064171122994653, | |
| "grad_norm": 34.769287109375, | |
| "learning_rate": 6.8400000000000014e-06, | |
| "loss": 0.0438, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.278074866310161, | |
| "grad_norm": 25.370189666748047, | |
| "learning_rate": 7.24e-06, | |
| "loss": 0.0541, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.491978609625669, | |
| "grad_norm": 32.902244567871094, | |
| "learning_rate": 7.640000000000001e-06, | |
| "loss": 0.1416, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 47.98198699951172, | |
| "learning_rate": 8.040000000000001e-06, | |
| "loss": 0.0939, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.919786096256685, | |
| "grad_norm": 16.503026962280273, | |
| "learning_rate": 8.44e-06, | |
| "loss": 0.1332, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.133689839572193, | |
| "grad_norm": 6.1169633865356445, | |
| "learning_rate": 8.84e-06, | |
| "loss": 0.0508, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.347593582887701, | |
| "grad_norm": 6.8135552406311035, | |
| "learning_rate": 9.240000000000001e-06, | |
| "loss": 0.0915, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.561497326203209, | |
| "grad_norm": 42.288963317871094, | |
| "learning_rate": 9.640000000000001e-06, | |
| "loss": 0.1364, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.775401069518717, | |
| "grad_norm": 68.40443420410156, | |
| "learning_rate": 1e-05, | |
| "loss": 0.183, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.989304812834225, | |
| "grad_norm": 18.801176071166992, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1234, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 6.2032085561497325, | |
| "grad_norm": 17.309553146362305, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0899, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 6.4171122994652405, | |
| "grad_norm": 39.96405792236328, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1682, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.6310160427807485, | |
| "grad_norm": 37.08085250854492, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1817, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 6.8449197860962565, | |
| "grad_norm": 41.293277740478516, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1853, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 19.117578506469727, | |
| "learning_rate": 1e-05, | |
| "loss": 0.084, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 3.196685552597046, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0796, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 7.4866310160427805, | |
| "grad_norm": 34.18360900878906, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0638, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.7005347593582885, | |
| "grad_norm": 34.174346923828125, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0499, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 7.9144385026737964, | |
| "grad_norm": 30.022125244140625, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0511, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 8.128342245989305, | |
| "grad_norm": 5.687129020690918, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0381, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 8.342245989304812, | |
| "grad_norm": 8.238314628601074, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0699, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 8.556149732620321, | |
| "grad_norm": 48.88713836669922, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1932, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.556149732620321, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 4.55859375, | |
| "eval_runtime": 0.8621, | |
| "eval_samples_per_second": 11.599, | |
| "eval_steps_per_second": 1.16, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 55, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.3017478812532736e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |