| { | |
| "best_metric": 0.701433002948761, | |
| "best_model_checkpoint": "vit-msn-small-beta-fia-manually-enhanced_test_1/checkpoint-26", | |
| "epoch": 28.571428571428573, | |
| "eval_steps": 500, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5714285714285714, | |
| "eval_accuracy": 0.3732394366197183, | |
| "eval_loss": 0.9554320573806763, | |
| "eval_runtime": 0.5997, | |
| "eval_samples_per_second": 236.775, | |
| "eval_steps_per_second": 5.002, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "eval_accuracy": 0.5211267605633803, | |
| "eval_loss": 0.9304441213607788, | |
| "eval_runtime": 0.662, | |
| "eval_samples_per_second": 214.511, | |
| "eval_steps_per_second": 4.532, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.6338028169014085, | |
| "eval_loss": 0.8979224562644958, | |
| "eval_runtime": 0.5817, | |
| "eval_samples_per_second": 244.096, | |
| "eval_steps_per_second": 5.157, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.647887323943662, | |
| "eval_loss": 0.8770471811294556, | |
| "eval_runtime": 0.6733, | |
| "eval_samples_per_second": 210.905, | |
| "eval_steps_per_second": 4.456, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "eval_accuracy": 0.6408450704225352, | |
| "eval_loss": 0.8595245480537415, | |
| "eval_runtime": 0.5929, | |
| "eval_samples_per_second": 239.513, | |
| "eval_steps_per_second": 5.06, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 3.3225924968719482, | |
| "learning_rate": 1e-05, | |
| "loss": 0.877, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.8456438183784485, | |
| "eval_runtime": 0.5986, | |
| "eval_samples_per_second": 237.231, | |
| "eval_steps_per_second": 5.012, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.4859154929577465, | |
| "eval_loss": 0.8825032711029053, | |
| "eval_runtime": 0.6716, | |
| "eval_samples_per_second": 211.445, | |
| "eval_steps_per_second": 4.467, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5563380281690141, | |
| "eval_loss": 0.8293709754943848, | |
| "eval_runtime": 0.647, | |
| "eval_samples_per_second": 219.459, | |
| "eval_steps_per_second": 4.636, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_accuracy": 0.6197183098591549, | |
| "eval_loss": 0.7883467674255371, | |
| "eval_runtime": 0.5861, | |
| "eval_samples_per_second": 242.294, | |
| "eval_steps_per_second": 5.119, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "eval_accuracy": 0.6549295774647887, | |
| "eval_loss": 0.7541125416755676, | |
| "eval_runtime": 0.5944, | |
| "eval_samples_per_second": 238.88, | |
| "eval_steps_per_second": 5.047, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.6690140845070423, | |
| "eval_loss": 0.7689406275749207, | |
| "eval_runtime": 0.5748, | |
| "eval_samples_per_second": 247.038, | |
| "eval_steps_per_second": 5.219, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 4.331999778747559, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.7053, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6619718309859155, | |
| "eval_loss": 0.765151858329773, | |
| "eval_runtime": 0.5898, | |
| "eval_samples_per_second": 240.78, | |
| "eval_steps_per_second": 5.087, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "eval_accuracy": 0.6690140845070423, | |
| "eval_loss": 0.7495813369750977, | |
| "eval_runtime": 0.6195, | |
| "eval_samples_per_second": 229.228, | |
| "eval_steps_per_second": 4.843, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "eval_accuracy": 0.7183098591549296, | |
| "eval_loss": 0.7138803005218506, | |
| "eval_runtime": 0.6203, | |
| "eval_samples_per_second": 228.92, | |
| "eval_steps_per_second": 4.836, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.7464788732394366, | |
| "eval_loss": 0.701433002948761, | |
| "eval_runtime": 0.6693, | |
| "eval_samples_per_second": 212.167, | |
| "eval_steps_per_second": 4.482, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7183098591549296, | |
| "eval_loss": 0.7289509773254395, | |
| "eval_runtime": 0.6002, | |
| "eval_samples_per_second": 236.581, | |
| "eval_steps_per_second": 4.998, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 16.571428571428573, | |
| "eval_accuracy": 0.6901408450704225, | |
| "eval_loss": 0.7431067824363708, | |
| "eval_runtime": 0.5779, | |
| "eval_samples_per_second": 245.72, | |
| "eval_steps_per_second": 5.191, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 9.143027305603027, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6176, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 17.714285714285715, | |
| "eval_accuracy": 0.6690140845070423, | |
| "eval_loss": 0.7498267889022827, | |
| "eval_runtime": 0.6748, | |
| "eval_samples_per_second": 210.435, | |
| "eval_steps_per_second": 4.446, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.676056338028169, | |
| "eval_loss": 0.7438530325889587, | |
| "eval_runtime": 0.5998, | |
| "eval_samples_per_second": 236.755, | |
| "eval_steps_per_second": 5.002, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6971830985915493, | |
| "eval_loss": 0.7346609234809875, | |
| "eval_runtime": 0.6421, | |
| "eval_samples_per_second": 221.132, | |
| "eval_steps_per_second": 4.672, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 20.571428571428573, | |
| "eval_accuracy": 0.6901408450704225, | |
| "eval_loss": 0.7377016544342041, | |
| "eval_runtime": 0.6794, | |
| "eval_samples_per_second": 209.017, | |
| "eval_steps_per_second": 4.416, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 21.714285714285715, | |
| "eval_accuracy": 0.6901408450704225, | |
| "eval_loss": 0.7227413058280945, | |
| "eval_runtime": 0.6118, | |
| "eval_samples_per_second": 232.105, | |
| "eval_steps_per_second": 4.904, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 3.607974052429199, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.6053, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.704225352112676, | |
| "eval_loss": 0.7227721810340881, | |
| "eval_runtime": 0.6582, | |
| "eval_samples_per_second": 215.737, | |
| "eval_steps_per_second": 4.558, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.6901408450704225, | |
| "eval_loss": 0.7282090187072754, | |
| "eval_runtime": 0.6208, | |
| "eval_samples_per_second": 228.753, | |
| "eval_steps_per_second": 4.833, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 24.571428571428573, | |
| "eval_accuracy": 0.676056338028169, | |
| "eval_loss": 0.7363458275794983, | |
| "eval_runtime": 0.6048, | |
| "eval_samples_per_second": 234.795, | |
| "eval_steps_per_second": 4.96, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "eval_accuracy": 0.6830985915492958, | |
| "eval_loss": 0.743062436580658, | |
| "eval_runtime": 0.6737, | |
| "eval_samples_per_second": 210.781, | |
| "eval_steps_per_second": 4.453, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.6830985915492958, | |
| "eval_loss": 0.7449536323547363, | |
| "eval_runtime": 0.6105, | |
| "eval_samples_per_second": 232.604, | |
| "eval_steps_per_second": 4.914, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.676056338028169, | |
| "eval_loss": 0.7454614043235779, | |
| "eval_runtime": 0.6143, | |
| "eval_samples_per_second": 231.153, | |
| "eval_steps_per_second": 4.884, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 5.9063825607299805, | |
| "learning_rate": 0.0, | |
| "loss": 0.5926, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "eval_accuracy": 0.676056338028169, | |
| "eval_loss": 0.7446578741073608, | |
| "eval_runtime": 0.5938, | |
| "eval_samples_per_second": 239.132, | |
| "eval_steps_per_second": 5.052, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "step": 50, | |
| "total_flos": 2.389739118386135e+17, | |
| "train_loss": 0.6795549583435059, | |
| "train_runtime": 125.8684, | |
| "train_samples_per_second": 169.622, | |
| "train_steps_per_second": 0.397 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.389739118386135e+17, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |