| { | |
| "best_metric": 0.6910678148269653, | |
| "best_model_checkpoint": "bert_base_lda_100_v1_rte/checkpoint-40", | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 90, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.8062904477119446, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.7572, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.713447630405426, | |
| "eval_runtime": 0.1924, | |
| "eval_samples_per_second": 1439.88, | |
| "eval_steps_per_second": 10.396, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.4334803819656372, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.7126, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6923933625221252, | |
| "eval_runtime": 0.1913, | |
| "eval_samples_per_second": 1448.012, | |
| "eval_steps_per_second": 10.455, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.9001538753509521, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6911, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5379061371841155, | |
| "eval_loss": 0.6912652254104614, | |
| "eval_runtime": 0.2155, | |
| "eval_samples_per_second": 1285.155, | |
| "eval_steps_per_second": 9.279, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.8192620873451233, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6914, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6910678148269653, | |
| "eval_runtime": 0.1917, | |
| "eval_samples_per_second": 1444.602, | |
| "eval_steps_per_second": 10.43, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.557899236679077, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6774, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5090252707581228, | |
| "eval_loss": 0.7089632153511047, | |
| "eval_runtime": 0.1914, | |
| "eval_samples_per_second": 1447.292, | |
| "eval_steps_per_second": 10.45, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.7362091541290283, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6709, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.48014440433212996, | |
| "eval_loss": 0.7789549231529236, | |
| "eval_runtime": 0.2059, | |
| "eval_samples_per_second": 1345.393, | |
| "eval_steps_per_second": 9.714, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.2326953411102295, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.5927, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5234657039711191, | |
| "eval_loss": 0.823212206363678, | |
| "eval_runtime": 0.1901, | |
| "eval_samples_per_second": 1457.47, | |
| "eval_steps_per_second": 10.523, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.318673610687256, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.511, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5415162454873647, | |
| "eval_loss": 0.9742180705070496, | |
| "eval_runtime": 0.1994, | |
| "eval_samples_per_second": 1389.271, | |
| "eval_steps_per_second": 10.031, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 9.328105926513672, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.4169, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5126353790613718, | |
| "eval_loss": 1.0650445222854614, | |
| "eval_runtime": 0.192, | |
| "eval_samples_per_second": 1443.08, | |
| "eval_steps_per_second": 10.419, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "step": 90, | |
| "total_flos": 2948159375308800.0, | |
| "train_loss": 0.6356894334157308, | |
| "train_runtime": 61.1905, | |
| "train_samples_per_second": 2034.631, | |
| "train_steps_per_second": 8.171 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 500, | |
| "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": 2948159375308800.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |