| { | |
| "best_metric": 0.7291533946990967, | |
| "best_model_checkpoint": "finetune_results/omarmomen/structformer_s1_final_with_pos/mnli-mm/checkpoint-6000", | |
| "epoch": 3.695150115473441, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "eval_accuracy": 0.614417552947998, | |
| "eval_loss": 0.8566737771034241, | |
| "eval_runtime": 12.1288, | |
| "eval_samples_per_second": 518.105, | |
| "eval_steps_per_second": 64.804, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.884526558891455e-05, | |
| "loss": 0.9715, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.6376512050628662, | |
| "eval_loss": 0.8208665251731873, | |
| "eval_runtime": 12.1366, | |
| "eval_samples_per_second": 517.771, | |
| "eval_steps_per_second": 64.763, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.7690531177829104e-05, | |
| "loss": 0.8412, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_accuracy": 0.6712285280227661, | |
| "eval_loss": 0.7672932147979736, | |
| "eval_runtime": 12.1384, | |
| "eval_samples_per_second": 517.695, | |
| "eval_steps_per_second": 64.753, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.653579676674365e-05, | |
| "loss": 0.8026, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.6904837489128113, | |
| "eval_loss": 0.7384242415428162, | |
| "eval_runtime": 12.1361, | |
| "eval_samples_per_second": 517.794, | |
| "eval_steps_per_second": 64.765, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.53810623556582e-05, | |
| "loss": 0.7778, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.6818904876708984, | |
| "eval_loss": 0.7472041249275208, | |
| "eval_runtime": 12.1396, | |
| "eval_samples_per_second": 517.643, | |
| "eval_steps_per_second": 64.747, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_accuracy": 0.7019414305686951, | |
| "eval_loss": 0.7058711647987366, | |
| "eval_runtime": 12.1415, | |
| "eval_samples_per_second": 517.566, | |
| "eval_steps_per_second": 64.737, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.422632794457275e-05, | |
| "loss": 0.7107, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_accuracy": 0.6974856853485107, | |
| "eval_loss": 0.7071972489356995, | |
| "eval_runtime": 12.1307, | |
| "eval_samples_per_second": 518.025, | |
| "eval_steps_per_second": 64.794, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 4.30715935334873e-05, | |
| "loss": 0.6742, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_accuracy": 0.7054423689842224, | |
| "eval_loss": 0.702299177646637, | |
| "eval_runtime": 12.1415, | |
| "eval_samples_per_second": 517.563, | |
| "eval_steps_per_second": 64.737, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 4.1916859122401844e-05, | |
| "loss": 0.67, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_accuracy": 0.7143539190292358, | |
| "eval_loss": 0.6831963062286377, | |
| "eval_runtime": 12.1328, | |
| "eval_samples_per_second": 517.935, | |
| "eval_steps_per_second": 64.783, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 4.07621247113164e-05, | |
| "loss": 0.6684, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.7118077874183655, | |
| "eval_loss": 0.695106029510498, | |
| "eval_runtime": 12.1396, | |
| "eval_samples_per_second": 517.643, | |
| "eval_steps_per_second": 64.747, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_accuracy": 0.7141947746276855, | |
| "eval_loss": 0.7270315885543823, | |
| "eval_runtime": 12.134, | |
| "eval_samples_per_second": 517.885, | |
| "eval_steps_per_second": 64.777, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 3.960739030023095e-05, | |
| "loss": 0.6182, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_accuracy": 0.7051241397857666, | |
| "eval_loss": 0.7234853506088257, | |
| "eval_runtime": 12.1395, | |
| "eval_samples_per_second": 517.649, | |
| "eval_steps_per_second": 64.747, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 3.84526558891455e-05, | |
| "loss": 0.5444, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_accuracy": 0.7169000506401062, | |
| "eval_loss": 0.6947318911552429, | |
| "eval_runtime": 12.151, | |
| "eval_samples_per_second": 517.16, | |
| "eval_steps_per_second": 64.686, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 3.729792147806005e-05, | |
| "loss": 0.5451, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "eval_accuracy": 0.7250159382820129, | |
| "eval_loss": 0.701257050037384, | |
| "eval_runtime": 12.1376, | |
| "eval_samples_per_second": 517.728, | |
| "eval_steps_per_second": 64.757, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 3.61431870669746e-05, | |
| "loss": 0.5473, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_accuracy": 0.7291533946990967, | |
| "eval_loss": 0.6873072981834412, | |
| "eval_runtime": 12.137, | |
| "eval_samples_per_second": 517.754, | |
| "eval_steps_per_second": 64.76, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.7162635326385498, | |
| "eval_loss": 0.7030926942825317, | |
| "eval_runtime": 12.1357, | |
| "eval_samples_per_second": 517.813, | |
| "eval_steps_per_second": 64.768, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.498845265588915e-05, | |
| "loss": 0.5482, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "eval_accuracy": 0.7113303542137146, | |
| "eval_loss": 0.7891779541969299, | |
| "eval_runtime": 12.1379, | |
| "eval_samples_per_second": 517.715, | |
| "eval_steps_per_second": 64.756, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.38337182448037e-05, | |
| "loss": 0.4137, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_accuracy": 0.7176957130432129, | |
| "eval_loss": 0.7793703675270081, | |
| "eval_runtime": 12.1394, | |
| "eval_samples_per_second": 517.654, | |
| "eval_steps_per_second": 64.748, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 3.2678983833718243e-05, | |
| "loss": 0.4196, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "eval_accuracy": 0.7189688086509705, | |
| "eval_loss": 0.773220419883728, | |
| "eval_runtime": 12.1325, | |
| "eval_samples_per_second": 517.947, | |
| "eval_steps_per_second": 64.785, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 3.1524249422632794e-05, | |
| "loss": 0.4286, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_accuracy": 0.7173774838447571, | |
| "eval_loss": 0.7898668646812439, | |
| "eval_runtime": 12.1342, | |
| "eval_samples_per_second": 517.877, | |
| "eval_steps_per_second": 64.776, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "step": 8000, | |
| "total_flos": 8.013030200197632e+16, | |
| "train_loss": 0.6363458843231201, | |
| "train_runtime": 4859.8552, | |
| "train_samples_per_second": 534.543, | |
| "train_steps_per_second": 4.455 | |
| } | |
| ], | |
| "max_steps": 21650, | |
| "num_train_epochs": 10, | |
| "total_flos": 8.013030200197632e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |