| { | |
| "best_metric": 0.3780292570590973, | |
| "best_model_checkpoint": "/sphinx/u/culturebank/tiktok_data/controversial_data/models/combined_data_distilbert_4/checkpoint-210", | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 210, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 2.306892156600952, | |
| "learning_rate": 7.619047619047618e-06, | |
| "loss": 0.6843, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 1.6495659351348877, | |
| "learning_rate": 7.238095238095238e-06, | |
| "loss": 0.6717, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 1.9257926940917969, | |
| "learning_rate": 6.857142857142856e-06, | |
| "loss": 0.6709, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6314077377319336, | |
| "eval_recall": 0.5, | |
| "eval_runtime": 0.1157, | |
| "eval_samples_per_second": 1011.601, | |
| "eval_steps_per_second": 69.169, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 2.3859009742736816, | |
| "learning_rate": 6.476190476190476e-06, | |
| "loss": 0.6529, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 2.7452704906463623, | |
| "learning_rate": 6.0952380952380945e-06, | |
| "loss": 0.6063, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 2.4575912952423096, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.6315, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.417150497436523, | |
| "learning_rate": 5.333333333333333e-06, | |
| "loss": 0.5804, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5548444390296936, | |
| "eval_recall": 0.9, | |
| "eval_runtime": 0.1059, | |
| "eval_samples_per_second": 1104.4, | |
| "eval_steps_per_second": 75.515, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 2.3926234245300293, | |
| "learning_rate": 4.952380952380952e-06, | |
| "loss": 0.5411, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 3.461317539215088, | |
| "learning_rate": 4.571428571428571e-06, | |
| "loss": 0.5031, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "grad_norm": 4.521629333496094, | |
| "learning_rate": 4.19047619047619e-06, | |
| "loss": 0.5264, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.4601297676563263, | |
| "eval_recall": 0.9, | |
| "eval_runtime": 0.113, | |
| "eval_samples_per_second": 1035.499, | |
| "eval_steps_per_second": 70.803, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "grad_norm": 5.5885233879089355, | |
| "learning_rate": 3.809523809523809e-06, | |
| "loss": 0.4245, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "grad_norm": 4.529420852661133, | |
| "learning_rate": 3.428571428571428e-06, | |
| "loss": 0.4956, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "grad_norm": 3.9411473274230957, | |
| "learning_rate": 3.0476190476190473e-06, | |
| "loss": 0.4511, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 9.669978141784668, | |
| "learning_rate": 2.6666666666666664e-06, | |
| "loss": 0.4133, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.40984398126602173, | |
| "eval_recall": 0.9, | |
| "eval_runtime": 0.1049, | |
| "eval_samples_per_second": 1114.827, | |
| "eval_steps_per_second": 76.228, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "grad_norm": 4.2742390632629395, | |
| "learning_rate": 2.2857142857142856e-06, | |
| "loss": 0.382, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "grad_norm": 3.2808213233947754, | |
| "learning_rate": 1.9047619047619045e-06, | |
| "loss": 0.3756, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "grad_norm": 4.833192348480225, | |
| "learning_rate": 1.5238095238095236e-06, | |
| "loss": 0.3707, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.3795815408229828, | |
| "eval_recall": 0.8, | |
| "eval_runtime": 0.1097, | |
| "eval_samples_per_second": 1066.759, | |
| "eval_steps_per_second": 72.941, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "grad_norm": 4.624128818511963, | |
| "learning_rate": 1.1428571428571428e-06, | |
| "loss": 0.3365, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "grad_norm": 6.549154758453369, | |
| "learning_rate": 7.619047619047618e-07, | |
| "loss": 0.3851, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "grad_norm": 2.928457021713257, | |
| "learning_rate": 3.809523809523809e-07, | |
| "loss": 0.32, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 8.7202787399292, | |
| "learning_rate": 0.0, | |
| "loss": 0.3771, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.3780292570590973, | |
| "eval_recall": 0.9, | |
| "eval_runtime": 0.1053, | |
| "eval_samples_per_second": 1111.275, | |
| "eval_steps_per_second": 75.985, | |
| "step": 210 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 210, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 92894315659848.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |