| { | |
| "best_metric": 1.0962358713150024, | |
| "best_model_checkpoint": "distilbert_lda_50_v1_mnli/checkpoint-13806", | |
| "epoch": 14.0, | |
| "eval_steps": 500, | |
| "global_step": 21476, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.02029508352279663, | |
| "learning_rate": 0.00098, | |
| "loss": 1.1024, | |
| "step": 1534 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.0992883443832397, | |
| "eval_runtime": 4.1317, | |
| "eval_samples_per_second": 2375.563, | |
| "eval_steps_per_second": 9.439, | |
| "step": 1534 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.11782866716384888, | |
| "learning_rate": 0.00096, | |
| "loss": 1.0986, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.3544574630667346, | |
| "eval_loss": 1.0987216234207153, | |
| "eval_runtime": 4.138, | |
| "eval_samples_per_second": 2371.933, | |
| "eval_steps_per_second": 9.425, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.038195185363292694, | |
| "learning_rate": 0.00094, | |
| "loss": 1.0987, | |
| "step": 4602 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.0989333391189575, | |
| "eval_runtime": 4.1583, | |
| "eval_samples_per_second": 2360.317, | |
| "eval_steps_per_second": 9.379, | |
| "step": 4602 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.062206000089645386, | |
| "learning_rate": 0.00092, | |
| "loss": 1.0986, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.10163414478302, | |
| "eval_runtime": 4.0968, | |
| "eval_samples_per_second": 2395.767, | |
| "eval_steps_per_second": 9.52, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.07343439757823944, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 1.0985, | |
| "step": 7670 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.3544574630667346, | |
| "eval_loss": 1.0988649129867554, | |
| "eval_runtime": 4.2047, | |
| "eval_samples_per_second": 2334.293, | |
| "eval_steps_per_second": 9.275, | |
| "step": 7670 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.047856856137514114, | |
| "learning_rate": 0.00088, | |
| "loss": 1.0987, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.3544574630667346, | |
| "eval_loss": 1.0988649129867554, | |
| "eval_runtime": 4.1454, | |
| "eval_samples_per_second": 2367.71, | |
| "eval_steps_per_second": 9.408, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.0791020318865776, | |
| "learning_rate": 0.00086, | |
| "loss": 1.0985, | |
| "step": 10738 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0968025922775269, | |
| "eval_runtime": 4.0855, | |
| "eval_samples_per_second": 2402.388, | |
| "eval_steps_per_second": 9.546, | |
| "step": 10738 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.016667626798152924, | |
| "learning_rate": 0.00084, | |
| "loss": 1.0986, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0990766286849976, | |
| "eval_runtime": 4.2035, | |
| "eval_samples_per_second": 2334.953, | |
| "eval_steps_per_second": 9.278, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.006651591043919325, | |
| "learning_rate": 0.00082, | |
| "loss": 1.0988, | |
| "step": 13806 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.0962358713150024, | |
| "eval_runtime": 4.1757, | |
| "eval_samples_per_second": 2350.501, | |
| "eval_steps_per_second": 9.34, | |
| "step": 13806 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.045499037951231, | |
| "learning_rate": 0.0008, | |
| "loss": 1.0986, | |
| "step": 15340 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.0992168188095093, | |
| "eval_runtime": 4.1657, | |
| "eval_samples_per_second": 2356.14, | |
| "eval_steps_per_second": 9.362, | |
| "step": 15340 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.009605828672647476, | |
| "learning_rate": 0.0007800000000000001, | |
| "loss": 1.0986, | |
| "step": 16874 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.0990049839019775, | |
| "eval_runtime": 4.099, | |
| "eval_samples_per_second": 2394.463, | |
| "eval_steps_per_second": 9.514, | |
| "step": 16874 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.03646823763847351, | |
| "learning_rate": 0.00076, | |
| "loss": 1.0984, | |
| "step": 18408 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0990766286849976, | |
| "eval_runtime": 4.3346, | |
| "eval_samples_per_second": 2264.34, | |
| "eval_steps_per_second": 8.997, | |
| "step": 18408 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.09090559184551239, | |
| "learning_rate": 0.00074, | |
| "loss": 1.0985, | |
| "step": 19942 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.31818644931227713, | |
| "eval_loss": 1.0965192317962646, | |
| "eval_runtime": 4.2309, | |
| "eval_samples_per_second": 2319.831, | |
| "eval_steps_per_second": 9.218, | |
| "step": 19942 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.03515646234154701, | |
| "learning_rate": 0.0007199999999999999, | |
| "loss": 1.0988, | |
| "step": 21476 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.3273560876209883, | |
| "eval_loss": 1.0987216234207153, | |
| "eval_runtime": 4.0634, | |
| "eval_samples_per_second": 2415.463, | |
| "eval_steps_per_second": 9.598, | |
| "step": 21476 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "step": 21476, | |
| "total_flos": 3.6414798065552794e+17, | |
| "train_loss": 1.098874776140008, | |
| "train_runtime": 5133.237, | |
| "train_samples_per_second": 3825.091, | |
| "train_steps_per_second": 14.942 | |
| } | |
| ], | |
| "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": 3.6414798065552794e+17, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |