| { | |
| "best_metric": 1.5001204013824463, | |
| "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e5l510-l/checkpoint-5500", | |
| "epoch": 1.013264554163596, | |
| "eval_steps": 500, | |
| "global_step": 5500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09211495946941783, | |
| "grad_norm": 21.660480499267578, | |
| "learning_rate": 4.907885040530583e-10, | |
| "loss": 0.3974, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09211495946941783, | |
| "eval_loss": 1.5028706789016724, | |
| "eval_runtime": 74.0662, | |
| "eval_samples_per_second": 16.296, | |
| "eval_steps_per_second": 2.039, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18422991893883567, | |
| "grad_norm": 60.26338577270508, | |
| "learning_rate": 4.815770081061165e-10, | |
| "loss": 0.462, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.18422991893883567, | |
| "eval_loss": 1.5025668144226074, | |
| "eval_runtime": 75.1835, | |
| "eval_samples_per_second": 16.054, | |
| "eval_steps_per_second": 2.008, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2763448784082535, | |
| "grad_norm": 1.4199143648147583, | |
| "learning_rate": 4.723655121591747e-10, | |
| "loss": 0.5249, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2763448784082535, | |
| "eval_loss": 1.5022265911102295, | |
| "eval_runtime": 76.3442, | |
| "eval_samples_per_second": 15.81, | |
| "eval_steps_per_second": 1.978, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.36845983787767134, | |
| "grad_norm": 1.2705016136169434, | |
| "learning_rate": 4.6315401621223287e-10, | |
| "loss": 0.4642, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.36845983787767134, | |
| "eval_loss": 1.5019283294677734, | |
| "eval_runtime": 77.0512, | |
| "eval_samples_per_second": 15.665, | |
| "eval_steps_per_second": 1.96, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.46057479734708917, | |
| "grad_norm": 335.6837463378906, | |
| "learning_rate": 4.5394252026529114e-10, | |
| "loss": 0.5056, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.46057479734708917, | |
| "eval_loss": 1.5015931129455566, | |
| "eval_runtime": 76.8527, | |
| "eval_samples_per_second": 15.705, | |
| "eval_steps_per_second": 1.965, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.552689756816507, | |
| "grad_norm": 402.1738586425781, | |
| "learning_rate": 4.4473102431834935e-10, | |
| "loss": 0.4604, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.552689756816507, | |
| "eval_loss": 1.501311182975769, | |
| "eval_runtime": 76.7728, | |
| "eval_samples_per_second": 15.722, | |
| "eval_steps_per_second": 1.967, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6448047162859248, | |
| "grad_norm": 120.70919799804688, | |
| "learning_rate": 4.3551952837140756e-10, | |
| "loss": 0.4508, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6448047162859248, | |
| "eval_loss": 1.5010316371917725, | |
| "eval_runtime": 77.0074, | |
| "eval_samples_per_second": 15.674, | |
| "eval_steps_per_second": 1.961, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7369196757553427, | |
| "grad_norm": 103.2845687866211, | |
| "learning_rate": 4.2630803242446577e-10, | |
| "loss": 0.4674, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7369196757553427, | |
| "eval_loss": 1.5007821321487427, | |
| "eval_runtime": 76.7277, | |
| "eval_samples_per_second": 15.731, | |
| "eval_steps_per_second": 1.968, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8290346352247605, | |
| "grad_norm": 464.5989074707031, | |
| "learning_rate": 4.17096536477524e-10, | |
| "loss": 0.4435, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8290346352247605, | |
| "eval_loss": 1.500537633895874, | |
| "eval_runtime": 76.8855, | |
| "eval_samples_per_second": 15.699, | |
| "eval_steps_per_second": 1.964, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.9211495946941783, | |
| "grad_norm": 6.3550262451171875, | |
| "learning_rate": 4.078850405305822e-10, | |
| "loss": 0.4573, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9211495946941783, | |
| "eval_loss": 1.5003185272216797, | |
| "eval_runtime": 77.2072, | |
| "eval_samples_per_second": 15.633, | |
| "eval_steps_per_second": 1.956, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.013264554163596, | |
| "grad_norm": 0.6333007216453552, | |
| "learning_rate": 3.9867354458364045e-10, | |
| "loss": 0.4681, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.013264554163596, | |
| "eval_loss": 1.5001204013824463, | |
| "eval_runtime": 76.9383, | |
| "eval_samples_per_second": 15.688, | |
| "eval_steps_per_second": 1.963, | |
| "step": 5500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 27140, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 1979390448632520.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |