| { | |
| "best_metric": 0.155641108751297, | |
| "best_model_checkpoint": "/cronus_data/avirinchipur/implicit_motives/schone_pow_roba_large/final/checkpoint-1660", | |
| "epoch": 4.992481203007519, | |
| "eval_steps": 500, | |
| "global_step": 1660, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15037593984962405, | |
| "grad_norm": 7.807601451873779, | |
| "learning_rate": 9.698795180722893e-06, | |
| "loss": 0.4607, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3007518796992481, | |
| "grad_norm": 4.938343524932861, | |
| "learning_rate": 9.397590361445785e-06, | |
| "loss": 0.383, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.45112781954887216, | |
| "grad_norm": 4.443939208984375, | |
| "learning_rate": 9.096385542168675e-06, | |
| "loss": 0.3231, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6015037593984962, | |
| "grad_norm": 3.3444550037384033, | |
| "learning_rate": 8.795180722891567e-06, | |
| "loss": 0.3008, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7518796992481203, | |
| "grad_norm": 6.133885383605957, | |
| "learning_rate": 8.493975903614459e-06, | |
| "loss": 0.2952, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9022556390977443, | |
| "grad_norm": 4.0399627685546875, | |
| "learning_rate": 8.19277108433735e-06, | |
| "loss": 0.2924, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9984962406015038, | |
| "eval_loss": 0.26054325699806213, | |
| "eval_roc_auc": 0.9248130559023198, | |
| "eval_runtime": 37.9704, | |
| "eval_samples_per_second": 429.309, | |
| "eval_steps_per_second": 1.686, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "grad_norm": 3.2617785930633545, | |
| "learning_rate": 7.891566265060243e-06, | |
| "loss": 0.2848, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.2030075187969924, | |
| "grad_norm": 2.823629140853882, | |
| "learning_rate": 7.590361445783133e-06, | |
| "loss": 0.2591, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3533834586466165, | |
| "grad_norm": 8.734086990356445, | |
| "learning_rate": 7.289156626506025e-06, | |
| "loss": 0.2651, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.5037593984962405, | |
| "grad_norm": 5.4109601974487305, | |
| "learning_rate": 6.987951807228917e-06, | |
| "loss": 0.2658, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.6541353383458648, | |
| "grad_norm": 3.0360822677612305, | |
| "learning_rate": 6.686746987951808e-06, | |
| "loss": 0.2609, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.8045112781954886, | |
| "grad_norm": 4.336048603057861, | |
| "learning_rate": 6.385542168674699e-06, | |
| "loss": 0.2629, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.954887218045113, | |
| "grad_norm": 3.214564800262451, | |
| "learning_rate": 6.084337349397591e-06, | |
| "loss": 0.2553, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.2233881801366806, | |
| "eval_roc_auc": 0.9468388936522675, | |
| "eval_runtime": 37.4563, | |
| "eval_samples_per_second": 435.2, | |
| "eval_steps_per_second": 1.709, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 3.9632742404937744, | |
| "learning_rate": 5.783132530120482e-06, | |
| "loss": 0.2392, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.255639097744361, | |
| "grad_norm": 3.3193111419677734, | |
| "learning_rate": 5.4819277108433745e-06, | |
| "loss": 0.2336, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.406015037593985, | |
| "grad_norm": 3.099203109741211, | |
| "learning_rate": 5.180722891566266e-06, | |
| "loss": 0.2195, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.556390977443609, | |
| "grad_norm": 5.596978664398193, | |
| "learning_rate": 4.8795180722891575e-06, | |
| "loss": 0.2339, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.706766917293233, | |
| "grad_norm": 3.742239236831665, | |
| "learning_rate": 4.578313253012049e-06, | |
| "loss": 0.2304, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 3.622905731201172, | |
| "learning_rate": 4.27710843373494e-06, | |
| "loss": 0.2317, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.998496240601504, | |
| "eval_loss": 0.189885213971138, | |
| "eval_roc_auc": 0.9619820442841185, | |
| "eval_runtime": 37.3548, | |
| "eval_samples_per_second": 436.383, | |
| "eval_steps_per_second": 1.713, | |
| "step": 997 | |
| }, | |
| { | |
| "epoch": 3.007518796992481, | |
| "grad_norm": 3.5725858211517334, | |
| "learning_rate": 3.975903614457832e-06, | |
| "loss": 0.229, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.1578947368421053, | |
| "grad_norm": 3.569227933883667, | |
| "learning_rate": 3.6746987951807235e-06, | |
| "loss": 0.1968, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.308270676691729, | |
| "grad_norm": 5.20859432220459, | |
| "learning_rate": 3.3734939759036146e-06, | |
| "loss": 0.2058, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.4586466165413534, | |
| "grad_norm": 6.604429244995117, | |
| "learning_rate": 3.072289156626506e-06, | |
| "loss": 0.2078, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.6090225563909772, | |
| "grad_norm": 4.930511474609375, | |
| "learning_rate": 2.771084337349398e-06, | |
| "loss": 0.2053, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.7593984962406015, | |
| "grad_norm": 3.622622489929199, | |
| "learning_rate": 2.469879518072289e-06, | |
| "loss": 0.2073, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.909774436090226, | |
| "grad_norm": 4.740933895111084, | |
| "learning_rate": 2.168674698795181e-06, | |
| "loss": 0.2063, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.16450706124305725, | |
| "eval_roc_auc": 0.9714800977158984, | |
| "eval_runtime": 37.3773, | |
| "eval_samples_per_second": 436.121, | |
| "eval_steps_per_second": 1.712, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 4.06015037593985, | |
| "grad_norm": 6.452109336853027, | |
| "learning_rate": 1.8674698795180723e-06, | |
| "loss": 0.1969, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.2105263157894735, | |
| "grad_norm": 4.532094478607178, | |
| "learning_rate": 1.566265060240964e-06, | |
| "loss": 0.1879, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.360902255639098, | |
| "grad_norm": 4.941292762756348, | |
| "learning_rate": 1.2650602409638555e-06, | |
| "loss": 0.1803, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.511278195488722, | |
| "grad_norm": 3.9667301177978516, | |
| "learning_rate": 9.638554216867472e-07, | |
| "loss": 0.1821, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.661654135338345, | |
| "grad_norm": 4.0561065673828125, | |
| "learning_rate": 6.626506024096387e-07, | |
| "loss": 0.1898, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 4.81203007518797, | |
| "grad_norm": 4.69659423828125, | |
| "learning_rate": 3.614457831325301e-07, | |
| "loss": 0.1885, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.962406015037594, | |
| "grad_norm": 4.419920444488525, | |
| "learning_rate": 6.02409638554217e-08, | |
| "loss": 0.1897, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 4.992481203007519, | |
| "eval_loss": 0.155641108751297, | |
| "eval_roc_auc": 0.9741751333869398, | |
| "eval_runtime": 37.3327, | |
| "eval_samples_per_second": 436.641, | |
| "eval_steps_per_second": 1.714, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 4.992481203007519, | |
| "step": 1660, | |
| "total_flos": 4.323837898114577e+16, | |
| "train_loss": 0.24427399233163122, | |
| "train_runtime": 2915.5625, | |
| "train_samples_per_second": 145.816, | |
| "train_steps_per_second": 0.569 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1660, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 4.323837898114577e+16, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |