| { | |
| "best_metric": 0.6522520780563354, | |
| "best_model_checkpoint": "bert_uncased_L-2_H-256_A-4_rte/checkpoint-80", | |
| "epoch": 13.0, | |
| "eval_steps": 500, | |
| "global_step": 130, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.49917563796043396, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6981, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5740072202166066, | |
| "eval_loss": 0.6831777691841125, | |
| "eval_runtime": 0.0995, | |
| "eval_samples_per_second": 2784.168, | |
| "eval_steps_per_second": 20.102, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.5291090607643127, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6877, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5740072202166066, | |
| "eval_loss": 0.6789401173591614, | |
| "eval_runtime": 0.1064, | |
| "eval_samples_per_second": 2603.336, | |
| "eval_steps_per_second": 18.797, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.6718031167984009, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6794, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5812274368231047, | |
| "eval_loss": 0.6745755076408386, | |
| "eval_runtime": 0.101, | |
| "eval_samples_per_second": 2741.744, | |
| "eval_steps_per_second": 19.796, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.6451788544654846, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6685, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5740072202166066, | |
| "eval_loss": 0.6702955365180969, | |
| "eval_runtime": 0.1088, | |
| "eval_samples_per_second": 2547.007, | |
| "eval_steps_per_second": 18.39, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.8607382774353027, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6592, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5848375451263538, | |
| "eval_loss": 0.6674469709396362, | |
| "eval_runtime": 0.1001, | |
| "eval_samples_per_second": 2767.996, | |
| "eval_steps_per_second": 19.986, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.7673319578170776, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6447, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6028880866425993, | |
| "eval_loss": 0.6636605858802795, | |
| "eval_runtime": 0.1128, | |
| "eval_samples_per_second": 2454.592, | |
| "eval_steps_per_second": 17.723, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.0504330396652222, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.6238, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5956678700361011, | |
| "eval_loss": 0.6564615368843079, | |
| "eval_runtime": 0.1002, | |
| "eval_samples_per_second": 2764.256, | |
| "eval_steps_per_second": 19.959, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.1754580736160278, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.6077, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6028880866425993, | |
| "eval_loss": 0.6522520780563354, | |
| "eval_runtime": 0.1019, | |
| "eval_samples_per_second": 2719.558, | |
| "eval_steps_per_second": 19.636, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.2603336572647095, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.5805, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5884476534296029, | |
| "eval_loss": 0.655805766582489, | |
| "eval_runtime": 0.1018, | |
| "eval_samples_per_second": 2721.156, | |
| "eval_steps_per_second": 19.647, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.229230523109436, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5502, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5848375451263538, | |
| "eval_loss": 0.6610111594200134, | |
| "eval_runtime": 0.0993, | |
| "eval_samples_per_second": 2790.293, | |
| "eval_steps_per_second": 20.147, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.8095612525939941, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.5119, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6064981949458483, | |
| "eval_loss": 0.6631617546081543, | |
| "eval_runtime": 0.0998, | |
| "eval_samples_per_second": 2775.522, | |
| "eval_steps_per_second": 20.04, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.673175573348999, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.4778, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6028880866425993, | |
| "eval_loss": 0.678740918636322, | |
| "eval_runtime": 0.0996, | |
| "eval_samples_per_second": 2779.991, | |
| "eval_steps_per_second": 20.072, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.942636251449585, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.4415, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.5956678700361011, | |
| "eval_loss": 0.7026975154876709, | |
| "eval_runtime": 0.1076, | |
| "eval_samples_per_second": 2573.991, | |
| "eval_steps_per_second": 18.585, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "step": 130, | |
| "total_flos": 81856452188160.0, | |
| "train_loss": 0.602387758401724, | |
| "train_runtime": 18.3661, | |
| "train_samples_per_second": 6778.809, | |
| "train_steps_per_second": 27.224 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 5 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 81856452188160.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |