| { | |
| "best_metric": 0.10110712051391602, | |
| "best_model_checkpoint": "/cronus_data/avirinchipur/implicit_motives/schone_aff_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": 3.783226251602173, | |
| "learning_rate": 9.698795180722893e-06, | |
| "loss": 0.3783, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3007518796992481, | |
| "grad_norm": 6.744827747344971, | |
| "learning_rate": 9.397590361445785e-06, | |
| "loss": 0.2572, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.45112781954887216, | |
| "grad_norm": 4.981093883514404, | |
| "learning_rate": 9.096385542168675e-06, | |
| "loss": 0.2298, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6015037593984962, | |
| "grad_norm": 3.8824877738952637, | |
| "learning_rate": 8.795180722891567e-06, | |
| "loss": 0.219, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7518796992481203, | |
| "grad_norm": 3.345761299133301, | |
| "learning_rate": 8.493975903614459e-06, | |
| "loss": 0.2094, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9022556390977443, | |
| "grad_norm": 5.345869541168213, | |
| "learning_rate": 8.19277108433735e-06, | |
| "loss": 0.2128, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9984962406015038, | |
| "eval_loss": 0.17600548267364502, | |
| "eval_roc_auc": 0.961318428605866, | |
| "eval_runtime": 37.7691, | |
| "eval_samples_per_second": 431.597, | |
| "eval_steps_per_second": 1.695, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "grad_norm": 4.458087921142578, | |
| "learning_rate": 7.891566265060243e-06, | |
| "loss": 0.2015, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.2030075187969924, | |
| "grad_norm": 8.24935531616211, | |
| "learning_rate": 7.590361445783133e-06, | |
| "loss": 0.1863, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3533834586466165, | |
| "grad_norm": 3.0867738723754883, | |
| "learning_rate": 7.289156626506025e-06, | |
| "loss": 0.1866, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.5037593984962405, | |
| "grad_norm": 3.783043384552002, | |
| "learning_rate": 6.987951807228917e-06, | |
| "loss": 0.1932, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.6541353383458648, | |
| "grad_norm": 3.2181010246276855, | |
| "learning_rate": 6.686746987951808e-06, | |
| "loss": 0.1877, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.8045112781954886, | |
| "grad_norm": 5.074103355407715, | |
| "learning_rate": 6.385542168674699e-06, | |
| "loss": 0.1871, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.954887218045113, | |
| "grad_norm": 3.3025379180908203, | |
| "learning_rate": 6.084337349397591e-06, | |
| "loss": 0.184, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.14929483830928802, | |
| "eval_roc_auc": 0.9729542423254111, | |
| "eval_runtime": 37.3448, | |
| "eval_samples_per_second": 436.5, | |
| "eval_steps_per_second": 1.714, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 6.176891326904297, | |
| "learning_rate": 5.783132530120482e-06, | |
| "loss": 0.174, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.255639097744361, | |
| "grad_norm": 3.600426435470581, | |
| "learning_rate": 5.4819277108433745e-06, | |
| "loss": 0.1559, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.406015037593985, | |
| "grad_norm": 4.919001579284668, | |
| "learning_rate": 5.180722891566266e-06, | |
| "loss": 0.1603, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.556390977443609, | |
| "grad_norm": 4.144702434539795, | |
| "learning_rate": 4.8795180722891575e-06, | |
| "loss": 0.1615, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.706766917293233, | |
| "grad_norm": 3.639554738998413, | |
| "learning_rate": 4.578313253012049e-06, | |
| "loss": 0.1712, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 3.972834348678589, | |
| "learning_rate": 4.27710843373494e-06, | |
| "loss": 0.1592, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.998496240601504, | |
| "eval_loss": 0.12567295134067535, | |
| "eval_roc_auc": 0.9811459479388318, | |
| "eval_runtime": 37.3949, | |
| "eval_samples_per_second": 435.915, | |
| "eval_steps_per_second": 1.711, | |
| "step": 997 | |
| }, | |
| { | |
| "epoch": 3.007518796992481, | |
| "grad_norm": 3.174755811691284, | |
| "learning_rate": 3.975903614457832e-06, | |
| "loss": 0.1656, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.1578947368421053, | |
| "grad_norm": 3.0254180431365967, | |
| "learning_rate": 3.6746987951807235e-06, | |
| "loss": 0.1377, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.308270676691729, | |
| "grad_norm": 3.0867421627044678, | |
| "learning_rate": 3.3734939759036146e-06, | |
| "loss": 0.1416, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.4586466165413534, | |
| "grad_norm": 4.309436321258545, | |
| "learning_rate": 3.072289156626506e-06, | |
| "loss": 0.147, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.6090225563909772, | |
| "grad_norm": 3.5912740230560303, | |
| "learning_rate": 2.771084337349398e-06, | |
| "loss": 0.1436, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.7593984962406015, | |
| "grad_norm": 5.856467247009277, | |
| "learning_rate": 2.469879518072289e-06, | |
| "loss": 0.1407, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.909774436090226, | |
| "grad_norm": 3.7223057746887207, | |
| "learning_rate": 2.168674698795181e-06, | |
| "loss": 0.1464, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.10692735016345978, | |
| "eval_roc_auc": 0.9861082356918816, | |
| "eval_runtime": 37.333, | |
| "eval_samples_per_second": 436.638, | |
| "eval_steps_per_second": 1.714, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 4.06015037593985, | |
| "grad_norm": 4.4520063400268555, | |
| "learning_rate": 1.8674698795180723e-06, | |
| "loss": 0.1344, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.2105263157894735, | |
| "grad_norm": 3.530614137649536, | |
| "learning_rate": 1.566265060240964e-06, | |
| "loss": 0.1348, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.360902255639098, | |
| "grad_norm": 6.2828474044799805, | |
| "learning_rate": 1.2650602409638555e-06, | |
| "loss": 0.127, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.511278195488722, | |
| "grad_norm": 5.152245044708252, | |
| "learning_rate": 9.638554216867472e-07, | |
| "loss": 0.1231, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.661654135338345, | |
| "grad_norm": 4.851140022277832, | |
| "learning_rate": 6.626506024096387e-07, | |
| "loss": 0.1273, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 4.81203007518797, | |
| "grad_norm": 4.454780101776123, | |
| "learning_rate": 3.614457831325301e-07, | |
| "loss": 0.1277, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.962406015037594, | |
| "grad_norm": 3.9713668823242188, | |
| "learning_rate": 6.02409638554217e-08, | |
| "loss": 0.1236, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 4.992481203007519, | |
| "eval_loss": 0.10110712051391602, | |
| "eval_roc_auc": 0.9877104952848222, | |
| "eval_runtime": 37.3953, | |
| "eval_samples_per_second": 435.91, | |
| "eval_steps_per_second": 1.711, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 4.992481203007519, | |
| "step": 1660, | |
| "total_flos": 4.323837898114577e+16, | |
| "train_loss": 0.17350520376699516, | |
| "train_runtime": 2909.4174, | |
| "train_samples_per_second": 146.124, | |
| "train_steps_per_second": 0.571 | |
| } | |
| ], | |
| "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 | |
| } | |