| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 2.4489102363586426, | |
| "learning_rate": 2e-05, | |
| "loss": 1.8047, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 1.8422876596450806, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 1.3832, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 3.1532328128814697, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.1404, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.9095581769943237, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 1.068, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 2.4410550594329834, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 1.014, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 2.7899439334869385, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 1.1168, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 2.6681783199310303, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.9987, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 2.2320525646209717, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 1.198, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 1.1953657865524292, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 1.0336, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.67486572265625, | |
| "learning_rate": 0.0, | |
| "loss": 1.0806, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 1.0614951848983765, | |
| "learning_rate": 1.913545457642601e-05, | |
| "loss": 0.9586, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.1396006345748901, | |
| "learning_rate": 1.8829475928589272e-05, | |
| "loss": 0.9822, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 1.8257782459259033, | |
| "learning_rate": 1.848048096156426e-05, | |
| "loss": 0.9798, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 3.6675124168395996, | |
| "learning_rate": 1.8090169943749477e-05, | |
| "loss": 0.9642, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.7332773208618164, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.0019, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 1.2114962339401245, | |
| "learning_rate": 1.7193398003386514e-05, | |
| "loss": 0.8493, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 5.460783004760742, | |
| "learning_rate": 1.6691306063588583e-05, | |
| "loss": 0.8319, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 3.551159620285034, | |
| "learning_rate": 1.6156614753256583e-05, | |
| "loss": 0.7986, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 4.315650463104248, | |
| "learning_rate": 1.5591929034707468e-05, | |
| "loss": 0.7908, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 5.731149673461914, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.8207, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "grad_norm": 2.3099749088287354, | |
| "learning_rate": 1.4383711467890776e-05, | |
| "loss": 0.8045, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 5.100424766540527, | |
| "learning_rate": 1.3746065934159123e-05, | |
| "loss": 0.6337, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "grad_norm": 4.819253444671631, | |
| "learning_rate": 1.3090169943749475e-05, | |
| "loss": 0.5244, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 5.7660675048828125, | |
| "learning_rate": 1.2419218955996677e-05, | |
| "loss": 0.5807, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.9891021251678467, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.6353, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "grad_norm": 3.6858391761779785, | |
| "learning_rate": 1.1045284632676535e-05, | |
| "loss": 0.4776, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "grad_norm": 9.160813331604004, | |
| "learning_rate": 1.0348994967025012e-05, | |
| "loss": 0.323, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "grad_norm": 7.0403313636779785, | |
| "learning_rate": 9.651005032974994e-06, | |
| "loss": 0.4799, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "grad_norm": 7.419447898864746, | |
| "learning_rate": 8.954715367323468e-06, | |
| "loss": 0.5816, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.421180009841919, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.4278, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "grad_norm": 4.215754985809326, | |
| "learning_rate": 7.580781044003324e-06, | |
| "loss": 0.3721, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "grad_norm": 6.3047308921813965, | |
| "learning_rate": 6.909830056250527e-06, | |
| "loss": 0.2973, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "grad_norm": 14.362757682800293, | |
| "learning_rate": 6.25393406584088e-06, | |
| "loss": 0.2589, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "grad_norm": 7.351701736450195, | |
| "learning_rate": 5.616288532109225e-06, | |
| "loss": 0.2718, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 8.06631851196289, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.3159, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "grad_norm": 5.070101261138916, | |
| "learning_rate": 4.408070965292534e-06, | |
| "loss": 0.2075, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "grad_norm": 7.613597869873047, | |
| "learning_rate": 3.8433852467434175e-06, | |
| "loss": 0.1164, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "grad_norm": 6.611764907836914, | |
| "learning_rate": 3.308693936411421e-06, | |
| "loss": 0.2217, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "grad_norm": 12.279210090637207, | |
| "learning_rate": 2.8066019966134907e-06, | |
| "loss": 0.1475, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 14.093499183654785, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.2752, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "grad_norm": 2.3949663639068604, | |
| "learning_rate": 1.9098300562505266e-06, | |
| "loss": 0.1019, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "grad_norm": 3.969956159591675, | |
| "learning_rate": 1.5195190384357405e-06, | |
| "loss": 0.1754, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "grad_norm": 8.477439880371094, | |
| "learning_rate": 1.1705240714107301e-06, | |
| "loss": 0.1174, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "grad_norm": 2.671515464782715, | |
| "learning_rate": 8.645454235739903e-07, | |
| "loss": 0.1461, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.29233822226524353, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.1105, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "grad_norm": 3.0615086555480957, | |
| "learning_rate": 3.8738304061681107e-07, | |
| "loss": 0.1009, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "grad_norm": 1.1887454986572266, | |
| "learning_rate": 2.1852399266194312e-07, | |
| "loss": 0.1134, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "grad_norm": 13.041033744812012, | |
| "learning_rate": 9.731931258429638e-08, | |
| "loss": 0.1092, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "grad_norm": 5.9538798332214355, | |
| "learning_rate": 2.4359497401758026e-08, | |
| "loss": 0.0799, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.9380784034729004, | |
| "learning_rate": 0.0, | |
| "loss": 0.1335, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 500, | |
| "total_flos": 1.1283387467931648e+16, | |
| "train_loss": 0.3623817192316055, | |
| "train_runtime": 5832.4394, | |
| "train_samples_per_second": 0.171, | |
| "train_steps_per_second": 0.086 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1283387467931648e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |