{ "best_metric": 1.3007307052612305, "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l56-l/checkpoint-3000", "epoch": 0.552689756816507, "eval_steps": 500, "global_step": 3000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.09211495946941783, "grad_norm": 18.028329849243164, "learning_rate": 4.846475067550971e-06, "loss": 0.2253, "step": 500 }, { "epoch": 0.09211495946941783, "eval_loss": 1.4693936109542847, "eval_runtime": 74.2802, "eval_samples_per_second": 16.249, "eval_steps_per_second": 2.033, "step": 500 }, { "epoch": 0.18422991893883567, "grad_norm": 0.4609026312828064, "learning_rate": 4.692950135101941e-06, "loss": 0.2544, "step": 1000 }, { "epoch": 0.18422991893883567, "eval_loss": 1.5420819520950317, "eval_runtime": 74.0766, "eval_samples_per_second": 16.294, "eval_steps_per_second": 2.038, "step": 1000 }, { "epoch": 0.2763448784082535, "grad_norm": 71.96737670898438, "learning_rate": 4.5394252026529115e-06, "loss": 0.3236, "step": 1500 }, { "epoch": 0.2763448784082535, "eval_loss": 1.4642689228057861, "eval_runtime": 74.2336, "eval_samples_per_second": 16.259, "eval_steps_per_second": 2.034, "step": 1500 }, { "epoch": 0.36845983787767134, "grad_norm": 6.366127069945549e-10, "learning_rate": 4.385900270203882e-06, "loss": 0.1888, "step": 2000 }, { "epoch": 0.36845983787767134, "eval_loss": 1.3193777799606323, "eval_runtime": 76.1214, "eval_samples_per_second": 15.856, "eval_steps_per_second": 1.984, "step": 2000 }, { "epoch": 0.46057479734708917, "grad_norm": 452.8532409667969, "learning_rate": 4.232375337754851e-06, "loss": 0.2563, "step": 2500 }, { "epoch": 0.46057479734708917, "eval_loss": 1.37644362449646, "eval_runtime": 76.3265, "eval_samples_per_second": 15.814, "eval_steps_per_second": 1.978, "step": 2500 }, { "epoch": 0.552689756816507, "grad_norm": 0.0002858435327652842, "learning_rate": 4.078850405305822e-06, "loss": 0.2794, "step": 3000 }, { "epoch": 0.552689756816507, "eval_loss": 1.3007307052612305, "eval_runtime": 76.3512, "eval_samples_per_second": 15.809, "eval_steps_per_second": 1.978, "step": 3000 } ], "logging_steps": 500, "max_steps": 16284, "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": 1079994570482520.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }