| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 41.0, | |
| "eval_steps": 500, | |
| "global_step": 56, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.581096649169922, | |
| "learning_rate": 6.25e-06, | |
| "loss": 1.2294, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.5810961723327637, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.2294, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.612244897959184, | |
| "grad_norm": 1.6523983478546143, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 0.7504, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.9309479594230652, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.4609, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.275179147720337, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.16, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.085817575454712, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.1088, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 5.224489795918367, | |
| "grad_norm": 0.6138802766799927, | |
| "learning_rate": 4.375e-05, | |
| "loss": 0.2543, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.3060435056686401, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7491, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.6941332817077637, | |
| "learning_rate": 4.9038461538461536e-05, | |
| "loss": 0.9088, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 7.836734693877551, | |
| "grad_norm": 1.4038004875183105, | |
| "learning_rate": 4.8076923076923084e-05, | |
| "loss": 0.7006, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.2940540313720703, | |
| "learning_rate": 4.711538461538462e-05, | |
| "loss": 0.1139, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.4644262790679932, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.682, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.4234392642974854, | |
| "learning_rate": 4.519230769230769e-05, | |
| "loss": 0.6219, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 10.448979591836734, | |
| "grad_norm": 0.5763714909553528, | |
| "learning_rate": 4.423076923076923e-05, | |
| "loss": 0.2485, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.6293026804924011, | |
| "learning_rate": 4.326923076923077e-05, | |
| "loss": 0.2973, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.0199288129806519, | |
| "learning_rate": 4.230769230769231e-05, | |
| "loss": 0.4975, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.9868380427360535, | |
| "learning_rate": 4.134615384615385e-05, | |
| "loss": 0.467, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 13.061224489795919, | |
| "grad_norm": 0.1089855432510376, | |
| "learning_rate": 4.038461538461539e-05, | |
| "loss": 0.0253, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.7933926582336426, | |
| "learning_rate": 3.942307692307692e-05, | |
| "loss": 0.3902, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.7683815956115723, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.393, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 15.673469387755102, | |
| "grad_norm": 0.5520622730255127, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2497, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.32301419973373413, | |
| "learning_rate": 3.653846153846154e-05, | |
| "loss": 0.1196, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.7165759205818176, | |
| "learning_rate": 3.557692307692308e-05, | |
| "loss": 0.3426, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.7063692212104797, | |
| "learning_rate": 3.461538461538462e-05, | |
| "loss": 0.3282, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 18.285714285714285, | |
| "grad_norm": 0.2430814504623413, | |
| "learning_rate": 3.365384615384616e-05, | |
| "loss": 0.0891, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.5586780309677124, | |
| "learning_rate": 3.269230769230769e-05, | |
| "loss": 0.2166, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.6888500452041626, | |
| "learning_rate": 3.1730769230769234e-05, | |
| "loss": 0.29, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 20.897959183673468, | |
| "grad_norm": 0.6251823902130127, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.2468, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "grad_norm": 0.13063542544841766, | |
| "learning_rate": 2.9807692307692308e-05, | |
| "loss": 0.0303, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 0.6627045273780823, | |
| "learning_rate": 2.8846153846153845e-05, | |
| "loss": 0.2563, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "grad_norm": 0.6512354612350464, | |
| "learning_rate": 2.7884615384615386e-05, | |
| "loss": 0.2464, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 23.510204081632654, | |
| "grad_norm": 0.3564887046813965, | |
| "learning_rate": 2.6923076923076923e-05, | |
| "loss": 0.1299, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.3812919557094574, | |
| "learning_rate": 2.5961538461538464e-05, | |
| "loss": 0.103, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 0.6363742351531982, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2202, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 0.6395919322967529, | |
| "learning_rate": 2.4038461538461542e-05, | |
| "loss": 0.2125, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 26.122448979591837, | |
| "grad_norm": 0.1440415233373642, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.0233, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "grad_norm": 0.5729821920394897, | |
| "learning_rate": 2.2115384615384616e-05, | |
| "loss": 0.176, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.6337006688117981, | |
| "learning_rate": 2.1153846153846154e-05, | |
| "loss": 0.1919, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 28.73469387755102, | |
| "grad_norm": 0.49902835488319397, | |
| "learning_rate": 2.0192307692307694e-05, | |
| "loss": 0.1374, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "grad_norm": 0.2010829895734787, | |
| "learning_rate": 1.923076923076923e-05, | |
| "loss": 0.047, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.6151719093322754, | |
| "learning_rate": 1.826923076923077e-05, | |
| "loss": 0.1743, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "grad_norm": 0.6123619079589844, | |
| "learning_rate": 1.730769230769231e-05, | |
| "loss": 0.1691, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 31.346938775510203, | |
| "grad_norm": 0.22967088222503662, | |
| "learning_rate": 1.6346153846153847e-05, | |
| "loss": 0.0646, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.4611968696117401, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.097, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "grad_norm": 0.6066409945487976, | |
| "learning_rate": 1.4423076923076923e-05, | |
| "loss": 0.1557, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 33.95918367346939, | |
| "grad_norm": 0.5780301094055176, | |
| "learning_rate": 1.3461538461538462e-05, | |
| "loss": 0.147, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "grad_norm": 0.07678268104791641, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0047, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 0.6018832325935364, | |
| "learning_rate": 1.153846153846154e-05, | |
| "loss": 0.1449, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 0.5997663140296936, | |
| "learning_rate": 1.0576923076923077e-05, | |
| "loss": 0.1419, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 36.57142857142857, | |
| "grad_norm": 0.3663066327571869, | |
| "learning_rate": 9.615384615384616e-06, | |
| "loss": 0.081, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "grad_norm": 0.2968898415565491, | |
| "learning_rate": 8.653846153846155e-06, | |
| "loss": 0.0571, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "grad_norm": 0.5936910510063171, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.1343, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "grad_norm": 0.5912988185882568, | |
| "learning_rate": 6.730769230769231e-06, | |
| "loss": 0.1323, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 39.183673469387756, | |
| "grad_norm": 0.1835460513830185, | |
| "learning_rate": 5.76923076923077e-06, | |
| "loss": 0.0188, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.4620586335659027, | |
| "learning_rate": 4.807692307692308e-06, | |
| "loss": 0.1104, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "grad_norm": 0.5855268239974976, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.1275, | |
| "step": 56 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 60, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 60, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2447556393402368e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |