| { | |
| "best_metric": 0.4631750285625458, | |
| "best_model_checkpoint": "./Vit-GPT2-COCO2017Flickr-85k-11/checkpoint-2500", | |
| "epoch": 0.4665484743864888, | |
| "eval_steps": 500, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09330969487729775, | |
| "grad_norm": 1.14389169216156, | |
| "learning_rate": 4.8444693293517484e-05, | |
| "loss": 0.378, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09330969487729775, | |
| "eval_gen_len": 11.7725, | |
| "eval_loss": 0.469292551279068, | |
| "eval_rouge1": 40.2274, | |
| "eval_rouge2": 15.0119, | |
| "eval_rougeL": 36.4563, | |
| "eval_rougeLsum": 36.4656, | |
| "eval_runtime": 450.3853, | |
| "eval_samples_per_second": 8.881, | |
| "eval_steps_per_second": 2.22, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1866193897545955, | |
| "grad_norm": 0.8436835408210754, | |
| "learning_rate": 4.6889386587034965e-05, | |
| "loss": 0.3748, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1866193897545955, | |
| "eval_gen_len": 12.16675, | |
| "eval_loss": 0.46400758624076843, | |
| "eval_rouge1": 40.199, | |
| "eval_rouge2": 15.321, | |
| "eval_rougeL": 36.4279, | |
| "eval_rougeLsum": 36.4457, | |
| "eval_runtime": 431.1423, | |
| "eval_samples_per_second": 9.278, | |
| "eval_steps_per_second": 2.319, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.27992908463189325, | |
| "grad_norm": 0.8730084896087646, | |
| "learning_rate": 4.5334079880552446e-05, | |
| "loss": 0.374, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.27992908463189325, | |
| "eval_gen_len": 11.8, | |
| "eval_loss": 0.4668748080730438, | |
| "eval_rouge1": 39.9523, | |
| "eval_rouge2": 15.0587, | |
| "eval_rougeL": 36.3639, | |
| "eval_rougeLsum": 36.375, | |
| "eval_runtime": 424.8986, | |
| "eval_samples_per_second": 9.414, | |
| "eval_steps_per_second": 2.354, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.373238779509191, | |
| "grad_norm": 0.8020262718200684, | |
| "learning_rate": 4.3778773174069934e-05, | |
| "loss": 0.3721, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.373238779509191, | |
| "eval_gen_len": 11.2095, | |
| "eval_loss": 0.46446970105171204, | |
| "eval_rouge1": 40.3597, | |
| "eval_rouge2": 15.2173, | |
| "eval_rougeL": 36.6938, | |
| "eval_rougeLsum": 36.705, | |
| "eval_runtime": 410.7504, | |
| "eval_samples_per_second": 9.738, | |
| "eval_steps_per_second": 2.435, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4665484743864888, | |
| "grad_norm": 0.8208432793617249, | |
| "learning_rate": 4.222346646758741e-05, | |
| "loss": 0.3673, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.4665484743864888, | |
| "eval_gen_len": 11.93425, | |
| "eval_loss": 0.4631750285625458, | |
| "eval_rouge1": 40.3875, | |
| "eval_rouge2": 15.2532, | |
| "eval_rougeL": 36.5923, | |
| "eval_rougeLsum": 36.6182, | |
| "eval_runtime": 441.242, | |
| "eval_samples_per_second": 9.065, | |
| "eval_steps_per_second": 2.266, | |
| "step": 2500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 16074, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.21855564480512e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |