| { |
| "best_global_step": 2844, |
| "best_metric": 0.39313429594039917, |
| "best_model_checkpoint": "distilbert_rand_50_v1_qqp/checkpoint-2844", |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 9954, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 2.4479265213012695, |
| "learning_rate": 4.9e-05, |
| "loss": 0.4792, |
| "step": 1422 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7869651249072471, |
| "eval_combined_score": 0.725829664100371, |
| "eval_f1": 0.6646942032934947, |
| "eval_loss": 0.4546143114566803, |
| "eval_runtime": 16.6725, |
| "eval_samples_per_second": 2424.955, |
| "eval_steps_per_second": 9.477, |
| "step": 1422 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 2.99049711227417, |
| "learning_rate": 4.8e-05, |
| "loss": 0.3719, |
| "step": 2844 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8182537719515212, |
| "eval_combined_score": 0.7852647054080308, |
| "eval_f1": 0.7522756388645405, |
| "eval_loss": 0.39313429594039917, |
| "eval_runtime": 16.4438, |
| "eval_samples_per_second": 2458.679, |
| "eval_steps_per_second": 9.608, |
| "step": 2844 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 4.110503673553467, |
| "learning_rate": 4.7e-05, |
| "loss": 0.2971, |
| "step": 4266 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8249319812020777, |
| "eval_combined_score": 0.7953193504156697, |
| "eval_f1": 0.7657067196292618, |
| "eval_loss": 0.4137588441371918, |
| "eval_runtime": 16.4167, |
| "eval_samples_per_second": 2462.731, |
| "eval_steps_per_second": 9.624, |
| "step": 4266 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 3.949686288833618, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.2387, |
| "step": 5688 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8321048726193421, |
| "eval_combined_score": 0.7984828090407228, |
| "eval_f1": 0.7648607454621034, |
| "eval_loss": 0.4155418574810028, |
| "eval_runtime": 16.3987, |
| "eval_samples_per_second": 2465.443, |
| "eval_steps_per_second": 9.635, |
| "step": 5688 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 1.9670673608779907, |
| "learning_rate": 4.5e-05, |
| "loss": 0.1924, |
| "step": 7110 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8351471679445956, |
| "eval_combined_score": 0.8034664105430904, |
| "eval_f1": 0.7717856531415853, |
| "eval_loss": 0.46950042247772217, |
| "eval_runtime": 16.2628, |
| "eval_samples_per_second": 2486.048, |
| "eval_steps_per_second": 9.715, |
| "step": 7110 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 3.5114402770996094, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.1575, |
| "step": 8532 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8350729656195894, |
| "eval_combined_score": 0.8026660307701035, |
| "eval_f1": 0.7702590959206174, |
| "eval_loss": 0.505837619304657, |
| "eval_runtime": 16.1985, |
| "eval_samples_per_second": 2495.915, |
| "eval_steps_per_second": 9.754, |
| "step": 8532 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 4.6916093826293945, |
| "learning_rate": 4.3e-05, |
| "loss": 0.1303, |
| "step": 9954 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8334899826861242, |
| "eval_combined_score": 0.806701837164953, |
| "eval_f1": 0.7799136916437819, |
| "eval_loss": 0.5472820401191711, |
| "eval_runtime": 16.2255, |
| "eval_samples_per_second": 2491.752, |
| "eval_steps_per_second": 9.738, |
| "step": 9954 |
| }, |
| { |
| "epoch": 7.0, |
| "step": 9954, |
| "total_flos": 1.686920659598684e+17, |
| "train_loss": 0.26674099171619137, |
| "train_runtime": 2427.168, |
| "train_samples_per_second": 7495.279, |
| "train_steps_per_second": 29.293 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 71100, |
| "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": 1.686920659598684e+17, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
