|
{ |
|
"best_metric": 0.395076185464859, |
|
"best_model_checkpoint": "mikhail-panzo/ceb_b128_le4_s4000/checkpoint-1500", |
|
"epoch": 158.41584158415841, |
|
"eval_steps": 500, |
|
"global_step": 2000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 3.9603960396039604, |
|
"grad_norm": 2.952680826187134, |
|
"learning_rate": 2.5e-06, |
|
"loss": 0.7707, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 7.920792079207921, |
|
"grad_norm": 1.126000165939331, |
|
"learning_rate": 5e-06, |
|
"loss": 0.6934, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 11.881188118811881, |
|
"grad_norm": 2.129329204559326, |
|
"learning_rate": 7.5e-06, |
|
"loss": 0.5982, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 15.841584158415841, |
|
"grad_norm": 1.1155015230178833, |
|
"learning_rate": 1e-05, |
|
"loss": 0.5093, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 19.801980198019802, |
|
"grad_norm": 1.3643360137939453, |
|
"learning_rate": 1.25e-05, |
|
"loss": 0.4912, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 23.762376237623762, |
|
"grad_norm": 0.9774149060249329, |
|
"learning_rate": 1.5e-05, |
|
"loss": 0.4764, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 27.722772277227723, |
|
"grad_norm": 0.8793525695800781, |
|
"learning_rate": 1.75e-05, |
|
"loss": 0.4648, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 31.683168316831683, |
|
"grad_norm": 0.7882798910140991, |
|
"learning_rate": 2e-05, |
|
"loss": 0.459, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 35.64356435643565, |
|
"grad_norm": 1.2033296823501587, |
|
"learning_rate": 2.25e-05, |
|
"loss": 0.4519, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 39.603960396039604, |
|
"grad_norm": 0.8620955348014832, |
|
"learning_rate": 2.5e-05, |
|
"loss": 0.4433, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 39.603960396039604, |
|
"eval_loss": 0.4056467115879059, |
|
"eval_runtime": 7.4199, |
|
"eval_samples_per_second": 24.259, |
|
"eval_steps_per_second": 3.1, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 43.56435643564357, |
|
"grad_norm": 1.266213297843933, |
|
"learning_rate": 2.7500000000000004e-05, |
|
"loss": 0.444, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 47.524752475247524, |
|
"grad_norm": 1.174134373664856, |
|
"learning_rate": 3e-05, |
|
"loss": 0.437, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 51.48514851485149, |
|
"grad_norm": 0.8188551664352417, |
|
"learning_rate": 3.2500000000000004e-05, |
|
"loss": 0.4318, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 55.445544554455445, |
|
"grad_norm": 1.3083854913711548, |
|
"learning_rate": 3.5e-05, |
|
"loss": 0.4297, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 59.40594059405941, |
|
"grad_norm": 0.8094576001167297, |
|
"learning_rate": 3.7500000000000003e-05, |
|
"loss": 0.4238, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 63.366336633663366, |
|
"grad_norm": 1.2506986856460571, |
|
"learning_rate": 4e-05, |
|
"loss": 0.4233, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 67.32673267326733, |
|
"grad_norm": 1.7902193069458008, |
|
"learning_rate": 4.25e-05, |
|
"loss": 0.4208, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 71.2871287128713, |
|
"grad_norm": 2.209643602371216, |
|
"learning_rate": 4.5e-05, |
|
"loss": 0.4195, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 75.24752475247524, |
|
"grad_norm": 2.256385087966919, |
|
"learning_rate": 4.75e-05, |
|
"loss": 0.4202, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 79.20792079207921, |
|
"grad_norm": 0.9167351126670837, |
|
"learning_rate": 5e-05, |
|
"loss": 0.4142, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 79.20792079207921, |
|
"eval_loss": 0.39529716968536377, |
|
"eval_runtime": 6.6198, |
|
"eval_samples_per_second": 27.191, |
|
"eval_steps_per_second": 3.474, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 83.16831683168317, |
|
"grad_norm": 0.9256124496459961, |
|
"learning_rate": 5.25e-05, |
|
"loss": 0.4137, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 87.12871287128714, |
|
"grad_norm": 1.9655492305755615, |
|
"learning_rate": 5.500000000000001e-05, |
|
"loss": 0.4132, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 91.08910891089108, |
|
"grad_norm": 1.08187997341156, |
|
"learning_rate": 5.7499999999999995e-05, |
|
"loss": 0.4078, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 95.04950495049505, |
|
"grad_norm": 1.8732513189315796, |
|
"learning_rate": 6e-05, |
|
"loss": 0.4061, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 99.00990099009901, |
|
"grad_norm": 1.3401920795440674, |
|
"learning_rate": 6.25e-05, |
|
"loss": 0.4054, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 102.97029702970298, |
|
"grad_norm": 0.8879286646842957, |
|
"learning_rate": 6.500000000000001e-05, |
|
"loss": 0.4046, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 106.93069306930693, |
|
"grad_norm": 1.2213553190231323, |
|
"learning_rate": 6.750000000000001e-05, |
|
"loss": 0.3999, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 110.89108910891089, |
|
"grad_norm": 2.05886173248291, |
|
"learning_rate": 7e-05, |
|
"loss": 0.3983, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 114.85148514851485, |
|
"grad_norm": 1.10814368724823, |
|
"learning_rate": 7.25e-05, |
|
"loss": 0.3964, |
|
"step": 1450 |
|
}, |
|
{ |
|
"epoch": 118.81188118811882, |
|
"grad_norm": 1.0220105648040771, |
|
"learning_rate": 7.500000000000001e-05, |
|
"loss": 0.3972, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 118.81188118811882, |
|
"eval_loss": 0.395076185464859, |
|
"eval_runtime": 7.1134, |
|
"eval_samples_per_second": 25.304, |
|
"eval_steps_per_second": 3.233, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 122.77227722772277, |
|
"grad_norm": 1.1671476364135742, |
|
"learning_rate": 7.75e-05, |
|
"loss": 0.3909, |
|
"step": 1550 |
|
}, |
|
{ |
|
"epoch": 126.73267326732673, |
|
"grad_norm": 0.8167665600776672, |
|
"learning_rate": 8e-05, |
|
"loss": 0.3913, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 130.69306930693068, |
|
"grad_norm": 1.9696727991104126, |
|
"learning_rate": 8.25e-05, |
|
"loss": 0.3888, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 134.65346534653466, |
|
"grad_norm": 1.1445192098617554, |
|
"learning_rate": 8.5e-05, |
|
"loss": 0.3885, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 138.6138613861386, |
|
"grad_norm": 1.908463954925537, |
|
"learning_rate": 8.75e-05, |
|
"loss": 0.3888, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 142.5742574257426, |
|
"grad_norm": 1.6358473300933838, |
|
"learning_rate": 9e-05, |
|
"loss": 0.3878, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 146.53465346534654, |
|
"grad_norm": 1.7407046556472778, |
|
"learning_rate": 9.250000000000001e-05, |
|
"loss": 0.3868, |
|
"step": 1850 |
|
}, |
|
{ |
|
"epoch": 150.4950495049505, |
|
"grad_norm": 3.984604597091675, |
|
"learning_rate": 9.5e-05, |
|
"loss": 0.3877, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 154.45544554455446, |
|
"grad_norm": 1.5218504667282104, |
|
"learning_rate": 9.75e-05, |
|
"loss": 0.3828, |
|
"step": 1950 |
|
}, |
|
{ |
|
"epoch": 158.41584158415841, |
|
"grad_norm": 1.2710033655166626, |
|
"learning_rate": 0.0001, |
|
"loss": 0.3806, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 158.41584158415841, |
|
"eval_loss": 0.39701494574546814, |
|
"eval_runtime": 7.067, |
|
"eval_samples_per_second": 25.47, |
|
"eval_steps_per_second": 3.255, |
|
"step": 2000 |
|
} |
|
], |
|
"logging_steps": 50, |
|
"max_steps": 4000, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 334, |
|
"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": 4.322114324975938e+16, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|