| { |
| "best_metric": 0.8621136553160071, |
| "best_model_checkpoint": "./nlu_finetuned_models/mrpc/roberta-base_lr1e-05/checkpoint-1656", |
| "epoch": 10.0, |
| "eval_steps": 500, |
| "global_step": 2070, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.776566757493188, |
| "eval_averaged_scores": 0.8158452162024243, |
| "eval_f1": 0.8551236749116607, |
| "eval_loss": 0.5290148258209229, |
| "eval_runtime": 0.2856, |
| "eval_samples_per_second": 1284.946, |
| "eval_steps_per_second": 80.528, |
| "step": 207 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8174386920980926, |
| "eval_averaged_scores": 0.8451519836202038, |
| "eval_f1": 0.872865275142315, |
| "eval_loss": 0.4343216121196747, |
| "eval_runtime": 0.268, |
| "eval_samples_per_second": 1369.624, |
| "eval_steps_per_second": 85.835, |
| "step": 414 |
| }, |
| { |
| "epoch": 2.4154589371980677, |
| "grad_norm": 36.1466064453125, |
| "learning_rate": 8.071979434447301e-06, |
| "loss": 0.46, |
| "step": 500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8092643051771117, |
| "eval_averaged_scores": 0.8354621921142475, |
| "eval_f1": 0.8616600790513834, |
| "eval_loss": 0.5129518508911133, |
| "eval_runtime": 0.2665, |
| "eval_samples_per_second": 1377.201, |
| "eval_steps_per_second": 86.31, |
| "step": 621 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8228882833787466, |
| "eval_averaged_scores": 0.8515914713394654, |
| "eval_f1": 0.8802946593001841, |
| "eval_loss": 0.5693352222442627, |
| "eval_runtime": 0.2664, |
| "eval_samples_per_second": 1377.754, |
| "eval_steps_per_second": 86.344, |
| "step": 828 |
| }, |
| { |
| "epoch": 4.830917874396135, |
| "grad_norm": 21.271873474121094, |
| "learning_rate": 5.501285347043702e-06, |
| "loss": 0.2384, |
| "step": 1000 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8147138964577657, |
| "eval_averaged_scores": 0.8401632723395548, |
| "eval_f1": 0.8656126482213439, |
| "eval_loss": 0.7265740036964417, |
| "eval_runtime": 0.2635, |
| "eval_samples_per_second": 1392.882, |
| "eval_steps_per_second": 87.292, |
| "step": 1035 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8228882833787466, |
| "eval_averaged_scores": 0.8478433589105083, |
| "eval_f1": 0.87279843444227, |
| "eval_loss": 0.7967910170555115, |
| "eval_runtime": 0.2652, |
| "eval_samples_per_second": 1383.735, |
| "eval_steps_per_second": 86.719, |
| "step": 1242 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8337874659400545, |
| "eval_averaged_scores": 0.8585763333524364, |
| "eval_f1": 0.8833652007648184, |
| "eval_loss": 0.9130034446716309, |
| "eval_runtime": 0.2653, |
| "eval_samples_per_second": 1383.206, |
| "eval_steps_per_second": 86.686, |
| "step": 1449 |
| }, |
| { |
| "epoch": 7.246376811594203, |
| "grad_norm": 0.06251997500658035, |
| "learning_rate": 2.9305912596401033e-06, |
| "loss": 0.1284, |
| "step": 1500 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8392370572207084, |
| "eval_averaged_scores": 0.8621136553160071, |
| "eval_f1": 0.884990253411306, |
| "eval_loss": 0.9644819498062134, |
| "eval_runtime": 0.2648, |
| "eval_samples_per_second": 1385.901, |
| "eval_steps_per_second": 86.855, |
| "step": 1656 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.8310626702997275, |
| "eval_averaged_scores": 0.8540233986419272, |
| "eval_f1": 0.876984126984127, |
| "eval_loss": 1.034262776374817, |
| "eval_runtime": 0.2652, |
| "eval_samples_per_second": 1383.894, |
| "eval_steps_per_second": 86.729, |
| "step": 1863 |
| }, |
| { |
| "epoch": 9.66183574879227, |
| "grad_norm": 0.12907934188842773, |
| "learning_rate": 3.598971722365039e-07, |
| "loss": 0.068, |
| "step": 2000 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.8310626702997275, |
| "eval_averaged_scores": 0.8552200511031711, |
| "eval_f1": 0.8793774319066148, |
| "eval_loss": 1.0211622714996338, |
| "eval_runtime": 0.2393, |
| "eval_samples_per_second": 1533.435, |
| "eval_steps_per_second": 96.101, |
| "step": 2070 |
| }, |
| { |
| "epoch": 10.0, |
| "step": 2070, |
| "total_flos": 1294854808964040.0, |
| "train_loss": 0.21819818883702374, |
| "train_runtime": 145.0146, |
| "train_samples_per_second": 227.632, |
| "train_steps_per_second": 14.274 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 2070, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 10, |
| "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": 1294854808964040.0, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
