| { | |
| "best_metric": 0.7575757575757576, | |
| "best_model_checkpoint": "dit-base-rvlcdip-finetuned-grp-actual/checkpoint-93", | |
| "epoch": 6.72, | |
| "eval_steps": 500, | |
| "global_step": 126, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 2.3577, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.5113636363636364, | |
| "eval_loss": 2.086306571960449, | |
| "eval_runtime": 240.3077, | |
| "eval_samples_per_second": 1.099, | |
| "eval_steps_per_second": 0.037, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.690265486725664e-05, | |
| "loss": 2.2163, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 4.247787610619469e-05, | |
| "loss": 2.0601, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.6477272727272727, | |
| "eval_loss": 1.8153679370880127, | |
| "eval_runtime": 14.6331, | |
| "eval_samples_per_second": 18.041, | |
| "eval_steps_per_second": 0.615, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 3.8053097345132744e-05, | |
| "loss": 1.943, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 3.3628318584070804e-05, | |
| "loss": 1.8068, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.6704545454545454, | |
| "eval_loss": 1.5881296396255493, | |
| "eval_runtime": 14.8318, | |
| "eval_samples_per_second": 17.8, | |
| "eval_steps_per_second": 0.607, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 2.9203539823008852e-05, | |
| "loss": 1.66, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 2.4778761061946905e-05, | |
| "loss": 1.5953, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7159090909090909, | |
| "eval_loss": 1.4111517667770386, | |
| "eval_runtime": 14.6707, | |
| "eval_samples_per_second": 17.995, | |
| "eval_steps_per_second": 0.613, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 2.0353982300884957e-05, | |
| "loss": 1.4929, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 1.592920353982301e-05, | |
| "loss": 1.4304, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "eval_accuracy": 0.7575757575757576, | |
| "eval_loss": 1.3033273220062256, | |
| "eval_runtime": 14.9724, | |
| "eval_samples_per_second": 17.632, | |
| "eval_steps_per_second": 0.601, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 1.1504424778761062e-05, | |
| "loss": 1.3606, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "learning_rate": 7.079646017699115e-06, | |
| "loss": 1.3458, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 1.2400753498077393, | |
| "eval_runtime": 14.8483, | |
| "eval_samples_per_second": 17.78, | |
| "eval_steps_per_second": 0.606, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 2.6548672566371683e-06, | |
| "loss": 1.3523, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_accuracy": 0.7575757575757576, | |
| "eval_loss": 1.2240339517593384, | |
| "eval_runtime": 14.9425, | |
| "eval_samples_per_second": 17.668, | |
| "eval_steps_per_second": 0.602, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "step": 126, | |
| "total_flos": 1.2388649195611423e+18, | |
| "train_loss": 1.7003454405163962, | |
| "train_runtime": 3667.1764, | |
| "train_samples_per_second": 4.535, | |
| "train_steps_per_second": 0.034 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 126, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 1.2388649195611423e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |