| { |
| "best_global_step": 6136, |
| "best_metric": 0.7837694883346558, |
| "best_model_checkpoint": "distilbert_km_10_v1_mnli/checkpoint-6136", |
| "epoch": 9.0, |
| "eval_steps": 500, |
| "global_step": 13806, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.0665395259857178, |
| "learning_rate": 4.9e-05, |
| "loss": 0.984, |
| "step": 1534 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.5617931737137035, |
| "eval_loss": 0.9150222539901733, |
| "eval_runtime": 4.0856, |
| "eval_samples_per_second": 2402.336, |
| "eval_steps_per_second": 9.546, |
| "step": 1534 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.649280309677124, |
| "learning_rate": 4.8e-05, |
| "loss": 0.8738, |
| "step": 3068 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.6174223127865512, |
| "eval_loss": 0.8359196782112122, |
| "eval_runtime": 3.9611, |
| "eval_samples_per_second": 2477.855, |
| "eval_steps_per_second": 9.846, |
| "step": 3068 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.967570185661316, |
| "learning_rate": 4.7e-05, |
| "loss": 0.7796, |
| "step": 4602 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.6512480896586856, |
| "eval_loss": 0.7962499260902405, |
| "eval_runtime": 3.9862, |
| "eval_samples_per_second": 2462.22, |
| "eval_steps_per_second": 9.784, |
| "step": 4602 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 2.134730815887451, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.6962, |
| "step": 6136 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.6634742740703006, |
| "eval_loss": 0.7837694883346558, |
| "eval_runtime": 3.9797, |
| "eval_samples_per_second": 2466.287, |
| "eval_steps_per_second": 9.8, |
| "step": 6136 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 2.4140524864196777, |
| "learning_rate": 4.5e-05, |
| "loss": 0.6204, |
| "step": 7670 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6658176260825267, |
| "eval_loss": 0.8045682311058044, |
| "eval_runtime": 3.9644, |
| "eval_samples_per_second": 2475.756, |
| "eval_steps_per_second": 9.837, |
| "step": 7670 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 2.6782379150390625, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.5434, |
| "step": 9204 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.6698930208863983, |
| "eval_loss": 0.8481786251068115, |
| "eval_runtime": 3.9769, |
| "eval_samples_per_second": 2468.015, |
| "eval_steps_per_second": 9.807, |
| "step": 9204 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 2.8815577030181885, |
| "learning_rate": 4.3e-05, |
| "loss": 0.468, |
| "step": 10738 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.667142129393785, |
| "eval_loss": 0.8989856243133545, |
| "eval_runtime": 3.9535, |
| "eval_samples_per_second": 2482.627, |
| "eval_steps_per_second": 9.865, |
| "step": 10738 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 3.2627060413360596, |
| "learning_rate": 4.2e-05, |
| "loss": 0.3994, |
| "step": 12272 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.655017829852267, |
| "eval_loss": 1.020479440689087, |
| "eval_runtime": 3.9542, |
| "eval_samples_per_second": 2482.143, |
| "eval_steps_per_second": 9.863, |
| "step": 12272 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 3.84536075592041, |
| "learning_rate": 4.1e-05, |
| "loss": 0.3374, |
| "step": 13806 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.6654100866021396, |
| "eval_loss": 1.1316611766815186, |
| "eval_runtime": 3.971, |
| "eval_samples_per_second": 2471.658, |
| "eval_steps_per_second": 9.821, |
| "step": 13806 |
| }, |
| { |
| "epoch": 9.0, |
| "step": 13806, |
| "total_flos": 2.340951304214108e+17, |
| "train_loss": 0.6335903069911859, |
| "train_runtime": 3242.4824, |
| "train_samples_per_second": 6055.576, |
| "train_steps_per_second": 23.655 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 76700, |
| "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": 2.340951304214108e+17, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
