| { | |
| "best_metric": 0.3048795163631439, | |
| "best_model_checkpoint": "./models/bert-italian-xxl-cased-ItaCoLA/checkpoint-700", | |
| "epoch": 10.0, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4243, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_accuracy": 0.854122621564482, | |
| "eval_loss": 0.38053709268569946, | |
| "eval_runtime": 3.6758, | |
| "eval_samples_per_second": 257.362, | |
| "eval_steps_per_second": 16.323, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3806, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_accuracy": 0.857293868921776, | |
| "eval_loss": 0.3861564099788666, | |
| "eval_runtime": 3.5095, | |
| "eval_samples_per_second": 269.553, | |
| "eval_steps_per_second": 17.096, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3289, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_accuracy": 0.8678646934460887, | |
| "eval_loss": 0.35367050766944885, | |
| "eval_runtime": 3.7169, | |
| "eval_samples_per_second": 254.517, | |
| "eval_steps_per_second": 16.143, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2967, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "eval_accuracy": 0.8826638477801269, | |
| "eval_loss": 0.3165006935596466, | |
| "eval_runtime": 3.5485, | |
| "eval_samples_per_second": 266.59, | |
| "eval_steps_per_second": 16.908, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2992, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_accuracy": 0.8784355179704016, | |
| "eval_loss": 0.33816611766815186, | |
| "eval_runtime": 3.7064, | |
| "eval_samples_per_second": 255.231, | |
| "eval_steps_per_second": 16.188, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2273, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "eval_accuracy": 0.8816067653276956, | |
| "eval_loss": 0.329441100358963, | |
| "eval_runtime": 3.7539, | |
| "eval_samples_per_second": 252.003, | |
| "eval_steps_per_second": 15.983, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2215, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "eval_accuracy": 0.8911205073995772, | |
| "eval_loss": 0.3048795163631439, | |
| "eval_runtime": 3.5485, | |
| "eval_samples_per_second": 266.591, | |
| "eval_steps_per_second": 16.909, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1718, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "eval_accuracy": 0.8911205073995772, | |
| "eval_loss": 0.35308167338371277, | |
| "eval_runtime": 3.56, | |
| "eval_samples_per_second": 265.727, | |
| "eval_steps_per_second": 16.854, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1757, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "eval_accuracy": 0.8921775898520085, | |
| "eval_loss": 0.39034464955329895, | |
| "eval_runtime": 3.7329, | |
| "eval_samples_per_second": 253.42, | |
| "eval_steps_per_second": 16.073, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1698, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "eval_accuracy": 0.8953488372093024, | |
| "eval_loss": 0.3871164321899414, | |
| "eval_runtime": 3.5471, | |
| "eval_samples_per_second": 266.695, | |
| "eval_steps_per_second": 16.915, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1307, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "eval_accuracy": 0.8953488372093024, | |
| "eval_loss": 0.42552220821380615, | |
| "eval_runtime": 3.5867, | |
| "eval_samples_per_second": 263.752, | |
| "eval_steps_per_second": 16.728, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1426, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "eval_accuracy": 0.8985200845665962, | |
| "eval_loss": 0.37290310859680176, | |
| "eval_runtime": 3.5578, | |
| "eval_samples_per_second": 265.898, | |
| "eval_steps_per_second": 16.865, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1136, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "eval_accuracy": 0.8964059196617337, | |
| "eval_loss": 0.49388667941093445, | |
| "eval_runtime": 3.5671, | |
| "eval_samples_per_second": 265.203, | |
| "eval_steps_per_second": 16.82, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1163, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "eval_accuracy": 0.8964059196617337, | |
| "eval_loss": 0.4004368782043457, | |
| "eval_runtime": 3.7365, | |
| "eval_samples_per_second": 253.18, | |
| "eval_steps_per_second": 16.058, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0936, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "eval_accuracy": 0.8964059196617337, | |
| "eval_loss": 0.5116058588027954, | |
| "eval_runtime": 3.5814, | |
| "eval_samples_per_second": 264.146, | |
| "eval_steps_per_second": 16.753, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0973, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "eval_accuracy": 0.8921775898520085, | |
| "eval_loss": 0.4807981848716736, | |
| "eval_runtime": 3.7354, | |
| "eval_samples_per_second": 253.253, | |
| "eval_steps_per_second": 16.063, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0899, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.8868921775898521, | |
| "eval_loss": 0.48126643896102905, | |
| "eval_runtime": 3.5551, | |
| "eval_samples_per_second": 266.094, | |
| "eval_steps_per_second": 16.877, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0687, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "eval_accuracy": 0.8847780126849895, | |
| "eval_loss": 0.6045676469802856, | |
| "eval_runtime": 3.7224, | |
| "eval_samples_per_second": 254.138, | |
| "eval_steps_per_second": 16.119, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0709, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "eval_accuracy": 0.8964059196617337, | |
| "eval_loss": 0.5939581394195557, | |
| "eval_runtime": 3.7591, | |
| "eval_samples_per_second": 251.655, | |
| "eval_steps_per_second": 15.961, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0694, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "eval_accuracy": 0.8911205073995772, | |
| "eval_loss": 0.5791226029396057, | |
| "eval_runtime": 3.5526, | |
| "eval_samples_per_second": 266.284, | |
| "eval_steps_per_second": 16.889, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0732, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "eval_accuracy": 0.8921775898520085, | |
| "eval_loss": 0.5576857924461365, | |
| "eval_runtime": 3.5693, | |
| "eval_samples_per_second": 265.038, | |
| "eval_steps_per_second": 16.81, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0714, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "eval_accuracy": 0.8995771670190275, | |
| "eval_loss": 0.5248907804489136, | |
| "eval_runtime": 3.7426, | |
| "eval_samples_per_second": 252.768, | |
| "eval_steps_per_second": 16.032, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0531, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "eval_accuracy": 0.8932346723044398, | |
| "eval_loss": 0.6098359227180481, | |
| "eval_runtime": 3.5458, | |
| "eval_samples_per_second": 266.794, | |
| "eval_steps_per_second": 16.921, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0713, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "eval_accuracy": 0.8942917547568711, | |
| "eval_loss": 0.5609742403030396, | |
| "eval_runtime": 3.5561, | |
| "eval_samples_per_second": 266.022, | |
| "eval_steps_per_second": 16.872, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2440, | |
| "total_flos": 5131323357158400.0, | |
| "train_loss": 0.16311937355604328, | |
| "train_runtime": 870.9833, | |
| "train_samples_per_second": 89.565, | |
| "train_steps_per_second": 2.801 | |
| } | |
| ], | |
| "max_steps": 2440, | |
| "num_train_epochs": 10, | |
| "total_flos": 5131323357158400.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |