| { | |
| "best_metric": 1.0927647352218628, | |
| "best_model_checkpoint": "/kaggle/output/checkpoint-12000", | |
| "epoch": 0.48891786179921776, | |
| "eval_steps": 1000, | |
| "global_step": 12000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 2.7777777777777777e-11, | |
| "loss": 1.2184, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.7750000000000004e-08, | |
| "loss": 1.1394, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_accuracy": 0.3327345309381238, | |
| "eval_loss": 1.1149410009384155, | |
| "eval_runtime": 20.6803, | |
| "eval_samples_per_second": 242.26, | |
| "eval_steps_per_second": 30.319, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 5.5527777777777784e-08, | |
| "loss": 1.1141, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.3401197604790419, | |
| "eval_loss": 1.104099988937378, | |
| "eval_runtime": 20.8477, | |
| "eval_samples_per_second": 240.314, | |
| "eval_steps_per_second": 30.075, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 8.330555555555556e-08, | |
| "loss": 1.116, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.3407185628742515, | |
| "eval_loss": 1.1040862798690796, | |
| "eval_runtime": 20.6818, | |
| "eval_samples_per_second": 242.242, | |
| "eval_steps_per_second": 30.317, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.1108333333333333e-07, | |
| "loss": 1.1158, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_accuracy": 0.32894211576846305, | |
| "eval_loss": 1.1020556688308716, | |
| "eval_runtime": 20.8541, | |
| "eval_samples_per_second": 240.241, | |
| "eval_steps_per_second": 30.066, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.3883333333333335e-07, | |
| "loss": 1.1135, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.34271457085828344, | |
| "eval_loss": 1.1008552312850952, | |
| "eval_runtime": 20.8055, | |
| "eval_samples_per_second": 240.802, | |
| "eval_steps_per_second": 30.136, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.6658333333333335e-07, | |
| "loss": 1.1121, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.3395209580838323, | |
| "eval_loss": 1.1004050970077515, | |
| "eval_runtime": 20.8985, | |
| "eval_samples_per_second": 239.731, | |
| "eval_steps_per_second": 30.002, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.9436111111111112e-07, | |
| "loss": 1.1089, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 0.35788423153692617, | |
| "eval_loss": 1.0985721349716187, | |
| "eval_runtime": 20.84, | |
| "eval_samples_per_second": 240.403, | |
| "eval_steps_per_second": 30.086, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 2.2213888888888891e-07, | |
| "loss": 1.1079, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.3331337325349301, | |
| "eval_loss": 1.098374843597412, | |
| "eval_runtime": 20.7886, | |
| "eval_samples_per_second": 240.998, | |
| "eval_steps_per_second": 30.161, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 2.4988888888888893e-07, | |
| "loss": 1.1087, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.34510978043912177, | |
| "eval_loss": 1.0993521213531494, | |
| "eval_runtime": 20.782, | |
| "eval_samples_per_second": 241.074, | |
| "eval_steps_per_second": 30.17, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 2.776666666666667e-07, | |
| "loss": 1.109, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_accuracy": 0.3475049900199601, | |
| "eval_loss": 1.0967597961425781, | |
| "eval_runtime": 20.6798, | |
| "eval_samples_per_second": 242.265, | |
| "eval_steps_per_second": 30.319, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.054444444444444e-07, | |
| "loss": 1.1052, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_accuracy": 0.37544910179640717, | |
| "eval_loss": 1.0941349267959595, | |
| "eval_runtime": 20.8641, | |
| "eval_samples_per_second": 240.126, | |
| "eval_steps_per_second": 30.052, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.3322222222222225e-07, | |
| "loss": 1.105, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_accuracy": 0.3834331337325349, | |
| "eval_loss": 1.0927647352218628, | |
| "eval_runtime": 20.6541, | |
| "eval_samples_per_second": 242.567, | |
| "eval_steps_per_second": 30.357, | |
| "step": 12000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 10000000, | |
| "num_train_epochs": 408, | |
| "save_steps": 1000, | |
| "total_flos": 8361873506304000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |