| { | |
| "best_metric": 1.0827423334121704, | |
| "best_model_checkpoint": "/kaggle/output/checkpoint-19000", | |
| "epoch": 0.7741199478487614, | |
| "eval_steps": 1000, | |
| "global_step": 19000, | |
| "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": 1.0722, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.7750000000000004e-08, | |
| "loss": 1.1407, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_accuracy": 0.32914171656686625, | |
| "eval_loss": 1.1048731803894043, | |
| "eval_runtime": 57.3781, | |
| "eval_samples_per_second": 87.316, | |
| "eval_steps_per_second": 10.928, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 5.5527777777777784e-08, | |
| "loss": 1.1253, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.3377245508982036, | |
| "eval_loss": 1.1033366918563843, | |
| "eval_runtime": 57.3237, | |
| "eval_samples_per_second": 87.398, | |
| "eval_steps_per_second": 10.938, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 8.327777777777778e-08, | |
| "loss": 1.1179, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.3502994011976048, | |
| "eval_loss": 1.104222059249878, | |
| "eval_runtime": 57.4527, | |
| "eval_samples_per_second": 87.202, | |
| "eval_steps_per_second": 10.913, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.1105555555555557e-07, | |
| "loss": 1.1197, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_accuracy": 0.3457085828343313, | |
| "eval_loss": 1.1000728607177734, | |
| "eval_runtime": 57.5257, | |
| "eval_samples_per_second": 87.091, | |
| "eval_steps_per_second": 10.899, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.3883333333333335e-07, | |
| "loss": 1.1202, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.3383233532934132, | |
| "eval_loss": 1.099381923675537, | |
| "eval_runtime": 57.825, | |
| "eval_samples_per_second": 86.641, | |
| "eval_steps_per_second": 10.843, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.6661111111111112e-07, | |
| "loss": 1.1166, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.35409181636726544, | |
| "eval_loss": 1.0975825786590576, | |
| "eval_runtime": 57.6509, | |
| "eval_samples_per_second": 86.902, | |
| "eval_steps_per_second": 10.876, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.9436111111111112e-07, | |
| "loss": 1.1126, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 0.35508982035928144, | |
| "eval_loss": 1.0957295894622803, | |
| "eval_runtime": 57.5423, | |
| "eval_samples_per_second": 87.066, | |
| "eval_steps_per_second": 10.896, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 2.2213888888888891e-07, | |
| "loss": 1.1091, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.36666666666666664, | |
| "eval_loss": 1.093665599822998, | |
| "eval_runtime": 57.2594, | |
| "eval_samples_per_second": 87.497, | |
| "eval_steps_per_second": 10.95, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 2.4988888888888893e-07, | |
| "loss": 1.1069, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.35129740518962077, | |
| "eval_loss": 1.1007317304611206, | |
| "eval_runtime": 57.5122, | |
| "eval_samples_per_second": 87.112, | |
| "eval_steps_per_second": 10.902, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 2.776666666666667e-07, | |
| "loss": 1.1054, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_accuracy": 0.38143712574850297, | |
| "eval_loss": 1.092604398727417, | |
| "eval_runtime": 57.5161, | |
| "eval_samples_per_second": 87.106, | |
| "eval_steps_per_second": 10.901, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.0541666666666667e-07, | |
| "loss": 1.1041, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_accuracy": 0.37704590818363276, | |
| "eval_loss": 1.0923383235931396, | |
| "eval_runtime": 57.5839, | |
| "eval_samples_per_second": 87.003, | |
| "eval_steps_per_second": 10.888, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.3319444444444444e-07, | |
| "loss": 1.1047, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_accuracy": 0.3914171656686627, | |
| "eval_loss": 1.0889469385147095, | |
| "eval_runtime": 57.877, | |
| "eval_samples_per_second": 86.563, | |
| "eval_steps_per_second": 10.833, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3.6094444444444446e-07, | |
| "loss": 1.1015, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_accuracy": 0.3437125748502994, | |
| "eval_loss": 1.095380187034607, | |
| "eval_runtime": 57.5214, | |
| "eval_samples_per_second": 87.098, | |
| "eval_steps_per_second": 10.9, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 3.8872222222222223e-07, | |
| "loss": 1.0997, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_accuracy": 0.3596806387225549, | |
| "eval_loss": 1.096056580543518, | |
| "eval_runtime": 57.4747, | |
| "eval_samples_per_second": 87.169, | |
| "eval_steps_per_second": 10.909, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.1650000000000006e-07, | |
| "loss": 1.1022, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.38882235528942116, | |
| "eval_loss": 1.089142084121704, | |
| "eval_runtime": 57.778, | |
| "eval_samples_per_second": 86.711, | |
| "eval_steps_per_second": 10.852, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.4427777777777783e-07, | |
| "loss": 1.0985, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.3930139720558882, | |
| "eval_loss": 1.0844948291778564, | |
| "eval_runtime": 57.5695, | |
| "eval_samples_per_second": 87.025, | |
| "eval_steps_per_second": 10.891, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.7202777777777785e-07, | |
| "loss": 1.0966, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_accuracy": 0.40459081836327343, | |
| "eval_loss": 1.083156704902649, | |
| "eval_runtime": 57.5224, | |
| "eval_samples_per_second": 87.096, | |
| "eval_steps_per_second": 10.9, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.998055555555556e-07, | |
| "loss": 1.0985, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_accuracy": 0.4055888223552894, | |
| "eval_loss": 1.0858113765716553, | |
| "eval_runtime": 57.413, | |
| "eval_samples_per_second": 87.263, | |
| "eval_steps_per_second": 10.921, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 5.275277777777778e-07, | |
| "loss": 1.0933, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_accuracy": 0.3996007984031936, | |
| "eval_loss": 1.0827423334121704, | |
| "eval_runtime": 57.2851, | |
| "eval_samples_per_second": 87.457, | |
| "eval_steps_per_second": 10.945, | |
| "step": 19000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 10000000, | |
| "num_train_epochs": 408, | |
| "save_steps": 1000, | |
| "total_flos": 3.9716747476992e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |