| { | |
| "best_metric": 0.6844189763069153, | |
| "best_model_checkpoint": "bert_tiny_lda_100_v1_wnli/checkpoint-18", | |
| "epoch": 11.0, | |
| "eval_steps": 500, | |
| "global_step": 33, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 8.367554664611816, | |
| "learning_rate": 0.00098, | |
| "loss": 0.988, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6867297291755676, | |
| "eval_runtime": 0.0295, | |
| "eval_samples_per_second": 2410.246, | |
| "eval_steps_per_second": 33.947, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.5282027125358582, | |
| "learning_rate": 0.00096, | |
| "loss": 0.694, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6856295466423035, | |
| "eval_runtime": 0.0289, | |
| "eval_samples_per_second": 2460.978, | |
| "eval_steps_per_second": 34.662, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.9888330101966858, | |
| "learning_rate": 0.00094, | |
| "loss": 0.7144, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6856295466423035, | |
| "eval_runtime": 0.0284, | |
| "eval_samples_per_second": 2497.971, | |
| "eval_steps_per_second": 35.183, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.2957957983016968, | |
| "learning_rate": 0.00092, | |
| "loss": 0.7011, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.6941022872924805, | |
| "eval_runtime": 0.0294, | |
| "eval_samples_per_second": 2412.218, | |
| "eval_steps_per_second": 33.975, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.28737398982048035, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 0.6932, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.7037302851676941, | |
| "eval_runtime": 0.0277, | |
| "eval_samples_per_second": 2565.434, | |
| "eval_steps_per_second": 36.133, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.4876857399940491, | |
| "learning_rate": 0.00088, | |
| "loss": 0.6964, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6844189763069153, | |
| "eval_runtime": 0.0275, | |
| "eval_samples_per_second": 2579.769, | |
| "eval_steps_per_second": 36.335, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.35543572902679443, | |
| "learning_rate": 0.00086, | |
| "loss": 0.6965, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6926164627075195, | |
| "eval_runtime": 0.0276, | |
| "eval_samples_per_second": 2574.861, | |
| "eval_steps_per_second": 36.266, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.9181599617004395, | |
| "learning_rate": 0.00084, | |
| "loss": 0.6957, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.7037302851676941, | |
| "eval_runtime": 0.0276, | |
| "eval_samples_per_second": 2572.304, | |
| "eval_steps_per_second": 36.23, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.4169580638408661, | |
| "learning_rate": 0.00082, | |
| "loss": 0.699, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.694597065448761, | |
| "eval_runtime": 0.0273, | |
| "eval_samples_per_second": 2599.813, | |
| "eval_steps_per_second": 36.617, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.704374372959137, | |
| "learning_rate": 0.0008, | |
| "loss": 0.6936, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6876101493835449, | |
| "eval_runtime": 0.0276, | |
| "eval_samples_per_second": 2573.526, | |
| "eval_steps_per_second": 36.247, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.26875758171081543, | |
| "learning_rate": 0.0007800000000000001, | |
| "loss": 0.6948, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.6944872736930847, | |
| "eval_runtime": 0.028, | |
| "eval_samples_per_second": 2538.427, | |
| "eval_steps_per_second": 35.752, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "step": 33, | |
| "total_flos": 183171693112320.0, | |
| "train_loss": 0.7242465091474128, | |
| "train_runtime": 11.5122, | |
| "train_samples_per_second": 2757.938, | |
| "train_steps_per_second": 13.03 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 150, | |
| "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": 183171693112320.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |