| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 13.0, | |
| "eval_steps": 500, | |
| "global_step": 65, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 5.582363636951145, | |
| "learning_rate": 1.4285714285714286e-06, | |
| "loss": 0.832, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 6.007906389380084, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8909, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 6.0264753137548634, | |
| "learning_rate": 4.2857142857142855e-06, | |
| "loss": 0.9083, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 4.758320667876842, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.877, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.4076577062917965, | |
| "learning_rate": 7.1428571428571436e-06, | |
| "loss": 0.7876, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 2.08237296256087, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.7694, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 4.6171372058213995, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8187, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 5.009434012692013, | |
| "learning_rate": 9.99266706925562e-06, | |
| "loss": 0.7953, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 6.9545840167778445, | |
| "learning_rate": 9.970689785771798e-06, | |
| "loss": 0.7978, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.75089354879985, | |
| "learning_rate": 9.934132612707631e-06, | |
| "loss": 0.7635, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 4.2068964184545585, | |
| "learning_rate": 9.883102778550434e-06, | |
| "loss": 0.7522, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 2.4112952356601105, | |
| "learning_rate": 9.817749962596115e-06, | |
| "loss": 0.6875, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 2.3091169308054056, | |
| "learning_rate": 9.738265855914014e-06, | |
| "loss": 0.6611, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 2.034937499528585, | |
| "learning_rate": 9.644883599083959e-06, | |
| "loss": 0.6835, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.6127281142448815, | |
| "learning_rate": 9.537877098354787e-06, | |
| "loss": 0.6588, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 1.2420963045502502, | |
| "learning_rate": 9.417560222230115e-06, | |
| "loss": 0.5886, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 1.3274360340575622, | |
| "learning_rate": 9.284285880837947e-06, | |
| "loss": 0.6684, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 1.0211469602843797, | |
| "learning_rate": 9.138444990784455e-06, | |
| "loss": 0.6007, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 0.9935488366512957, | |
| "learning_rate": 8.98046532852822e-06, | |
| "loss": 0.6003, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.8078724366399843, | |
| "learning_rate": 8.810810275638183e-06, | |
| "loss": 0.5275, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "grad_norm": 0.8588787511486784, | |
| "learning_rate": 8.629977459615655e-06, | |
| "loss": 0.5663, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 0.7538829403060613, | |
| "learning_rate": 8.438497294267117e-06, | |
| "loss": 0.5168, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "grad_norm": 0.7637932323734493, | |
| "learning_rate": 8.23693142390914e-06, | |
| "loss": 0.5235, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.7643306856468832, | |
| "learning_rate": 8.025871075968828e-06, | |
| "loss": 0.5304, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.6324846681806777, | |
| "learning_rate": 7.805935326811913e-06, | |
| "loss": 0.5152, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "grad_norm": 0.6902515181754407, | |
| "learning_rate": 7.57776928588511e-06, | |
| "loss": 0.5243, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "grad_norm": 0.6205487787202831, | |
| "learning_rate": 7.342042203498952e-06, | |
| "loss": 0.4908, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "grad_norm": 0.6281988958201551, | |
| "learning_rate": 7.099445507801324e-06, | |
| "loss": 0.4791, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "grad_norm": 0.6812591195712626, | |
| "learning_rate": 6.850690776699574e-06, | |
| "loss": 0.4376, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.5849169270560564, | |
| "learning_rate": 6.5965076506799e-06, | |
| "loss": 0.4264, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "grad_norm": 0.6457985712072969, | |
| "learning_rate": 6.337641692646106e-06, | |
| "loss": 0.4455, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "grad_norm": 0.5635801829905707, | |
| "learning_rate": 6.074852201055121e-06, | |
| "loss": 0.4173, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "grad_norm": 0.5223246354583466, | |
| "learning_rate": 5.808909982763825e-06, | |
| "loss": 0.401, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "grad_norm": 0.5498217637405548, | |
| "learning_rate": 5.540595092119709e-06, | |
| "loss": 0.4238, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.4720178214619839, | |
| "learning_rate": 5.270694542927089e-06, | |
| "loss": 0.3866, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "grad_norm": 0.47994799351007716, | |
| "learning_rate": 5e-06, | |
| "loss": 0.375, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "grad_norm": 0.4754551384737771, | |
| "learning_rate": 4.729305457072913e-06, | |
| "loss": 0.3888, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "grad_norm": 0.4812422271485711, | |
| "learning_rate": 4.459404907880293e-06, | |
| "loss": 0.3479, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "grad_norm": 0.45352582219328164, | |
| "learning_rate": 4.191090017236177e-06, | |
| "loss": 0.3618, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.4065323940483181, | |
| "learning_rate": 3.92514779894488e-06, | |
| "loss": 0.36, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "grad_norm": 0.4227442111084603, | |
| "learning_rate": 3.662358307353897e-06, | |
| "loss": 0.3046, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "grad_norm": 0.42414683307987505, | |
| "learning_rate": 3.403492349320101e-06, | |
| "loss": 0.3519, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "grad_norm": 0.40450567963312506, | |
| "learning_rate": 3.149309223300428e-06, | |
| "loss": 0.3273, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "grad_norm": 0.40272719775595717, | |
| "learning_rate": 2.9005544921986774e-06, | |
| "loss": 0.3316, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.41851005929072993, | |
| "learning_rate": 2.65795779650105e-06, | |
| "loss": 0.3278, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "grad_norm": 0.4311448019356393, | |
| "learning_rate": 2.422230714114891e-06, | |
| "loss": 0.3086, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "grad_norm": 0.35795131810925795, | |
| "learning_rate": 2.1940646731880887e-06, | |
| "loss": 0.2882, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "grad_norm": 0.386946992231837, | |
| "learning_rate": 1.9741289240311757e-06, | |
| "loss": 0.2919, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "grad_norm": 0.44062701543471006, | |
| "learning_rate": 1.7630685760908623e-06, | |
| "loss": 0.2985, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.4144356041601527, | |
| "learning_rate": 1.561502705732883e-06, | |
| "loss": 0.2974, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 10.2, | |
| "grad_norm": 0.3823205368468367, | |
| "learning_rate": 1.370022540384347e-06, | |
| "loss": 0.272, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "grad_norm": 0.40310131237975705, | |
| "learning_rate": 1.1891897243618184e-06, | |
| "loss": 0.296, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 10.6, | |
| "grad_norm": 0.3676392905535333, | |
| "learning_rate": 1.0195346714717813e-06, | |
| "loss": 0.2903, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 10.8, | |
| "grad_norm": 0.36362349004803485, | |
| "learning_rate": 8.615550092155478e-07, | |
| "loss": 0.2596, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.3644836984135043, | |
| "learning_rate": 7.157141191620548e-07, | |
| "loss": 0.2618, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "grad_norm": 0.3542459747594626, | |
| "learning_rate": 5.824397777698859e-07, | |
| "loss": 0.2768, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "grad_norm": 0.3647484821206581, | |
| "learning_rate": 4.6212290164521554e-07, | |
| "loss": 0.2579, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 11.6, | |
| "grad_norm": 0.3572146335428771, | |
| "learning_rate": 3.5511640091604293e-07, | |
| "loss": 0.2604, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 11.8, | |
| "grad_norm": 0.3263231215974691, | |
| "learning_rate": 2.617341440859883e-07, | |
| "loss": 0.2526, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.3500512368836611, | |
| "learning_rate": 1.8225003740388546e-07, | |
| "loss": 0.2719, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 12.2, | |
| "grad_norm": 0.3434723627022533, | |
| "learning_rate": 1.1689722144956672e-07, | |
| "loss": 0.2767, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 12.4, | |
| "grad_norm": 0.3535578464793507, | |
| "learning_rate": 6.58673872923693e-08, | |
| "loss": 0.2726, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 12.6, | |
| "grad_norm": 0.3308948888330182, | |
| "learning_rate": 2.9310214228202016e-08, | |
| "loss": 0.2583, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "grad_norm": 0.3515345904747637, | |
| "learning_rate": 7.332930744380906e-09, | |
| "loss": 0.262, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.33983888102120363, | |
| "learning_rate": 0.0, | |
| "loss": 0.2516, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "step": 65, | |
| "total_flos": 8.525099631326003e+16, | |
| "train_loss": 0.472858812717291, | |
| "train_runtime": 5413.2141, | |
| "train_samples_per_second": 0.759, | |
| "train_steps_per_second": 0.012 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 65, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 13, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.525099631326003e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |