| { |
| "best_global_step": 5126, |
| "best_metric": 0.14522501826286316, |
| "best_model_checkpoint": "bert_sentiment_model/checkpoint-5126", |
| "epoch": 2.0, |
| "eval_steps": 500, |
| "global_step": 10252, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.09754194303550527, |
| "grad_norm": 0.7001327872276306, |
| "learning_rate": 1.9351020939003774e-05, |
| "loss": 0.2293, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.19508388607101054, |
| "grad_norm": 4.297205924987793, |
| "learning_rate": 1.870074131876707e-05, |
| "loss": 0.1698, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.2926258291065158, |
| "grad_norm": 1.2200896739959717, |
| "learning_rate": 1.805046169853037e-05, |
| "loss": 0.1634, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.3901677721420211, |
| "grad_norm": 1.0850228071212769, |
| "learning_rate": 1.7400182078293668e-05, |
| "loss": 0.1635, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.4877097151775263, |
| "grad_norm": 0.29681870341300964, |
| "learning_rate": 1.6749902458056965e-05, |
| "loss": 0.1485, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.5852516582130316, |
| "grad_norm": 1.5155694484710693, |
| "learning_rate": 1.6099622837820262e-05, |
| "loss": 0.1566, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.6827936012485368, |
| "grad_norm": 0.7628584504127502, |
| "learning_rate": 1.5449343217583563e-05, |
| "loss": 0.1469, |
| "step": 3500 |
| }, |
| { |
| "epoch": 0.7803355442840422, |
| "grad_norm": 2.7121143341064453, |
| "learning_rate": 1.4799063597346861e-05, |
| "loss": 0.1467, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.8778774873195474, |
| "grad_norm": 4.958296298980713, |
| "learning_rate": 1.4148783977110158e-05, |
| "loss": 0.1512, |
| "step": 4500 |
| }, |
| { |
| "epoch": 0.9754194303550526, |
| "grad_norm": 3.2470967769622803, |
| "learning_rate": 1.3498504356873457e-05, |
| "loss": 0.1445, |
| "step": 5000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.9502548221122192, |
| "eval_loss": 0.14522501826286316, |
| "eval_runtime": 77.7071, |
| "eval_samples_per_second": 527.738, |
| "eval_steps_per_second": 16.498, |
| "step": 5126 |
| }, |
| { |
| "epoch": 1.0729613733905579, |
| "grad_norm": 1.8167076110839844, |
| "learning_rate": 1.2848224736636754e-05, |
| "loss": 0.1307, |
| "step": 5500 |
| }, |
| { |
| "epoch": 1.1705033164260632, |
| "grad_norm": 1.4334732294082642, |
| "learning_rate": 1.2197945116400054e-05, |
| "loss": 0.1273, |
| "step": 6000 |
| }, |
| { |
| "epoch": 1.2680452594615685, |
| "grad_norm": 3.130772590637207, |
| "learning_rate": 1.1547665496163351e-05, |
| "loss": 0.1185, |
| "step": 6500 |
| }, |
| { |
| "epoch": 1.3655872024970737, |
| "grad_norm": 1.2805956602096558, |
| "learning_rate": 1.089738587592665e-05, |
| "loss": 0.1254, |
| "step": 7000 |
| }, |
| { |
| "epoch": 1.463129145532579, |
| "grad_norm": 6.393189907073975, |
| "learning_rate": 1.0247106255689946e-05, |
| "loss": 0.1246, |
| "step": 7500 |
| }, |
| { |
| "epoch": 1.5606710885680841, |
| "grad_norm": 0.682952880859375, |
| "learning_rate": 9.596826635453245e-06, |
| "loss": 0.1213, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.6582130316035895, |
| "grad_norm": 3.8420190811157227, |
| "learning_rate": 8.946547015216544e-06, |
| "loss": 0.1242, |
| "step": 8500 |
| }, |
| { |
| "epoch": 1.7557549746390948, |
| "grad_norm": 5.53094482421875, |
| "learning_rate": 8.296267394979842e-06, |
| "loss": 0.1266, |
| "step": 9000 |
| }, |
| { |
| "epoch": 1.8532969176746001, |
| "grad_norm": 2.091581344604492, |
| "learning_rate": 7.64598777474314e-06, |
| "loss": 0.1296, |
| "step": 9500 |
| }, |
| { |
| "epoch": 1.9508388607101055, |
| "grad_norm": 1.8294498920440674, |
| "learning_rate": 6.995708154506439e-06, |
| "loss": 0.1175, |
| "step": 10000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.9524006925309079, |
| "eval_loss": 0.147334024310112, |
| "eval_runtime": 77.7359, |
| "eval_samples_per_second": 527.543, |
| "eval_steps_per_second": 16.492, |
| "step": 10252 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 15378, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "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": 1.0864639920144384e+16, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
