| { |
| "best_metric": 0.8411684097031022, |
| "best_model_checkpoint": "models/stance_tracker/checkpoint-5255", |
| "epoch": 5.0, |
| "eval_steps": 500, |
| "global_step": 5255, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.09514747859181731, |
| "grad_norm": 1180798.875, |
| "learning_rate": 1.9011406844106464e-05, |
| "loss": 4.8322, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.19029495718363462, |
| "grad_norm": 73790.9375, |
| "learning_rate": 3.802281368821293e-05, |
| "loss": 1.802, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.285442435775452, |
| "grad_norm": 106478.0546875, |
| "learning_rate": 5.703422053231939e-05, |
| "loss": 0.9777, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.38058991436726924, |
| "grad_norm": 246882.859375, |
| "learning_rate": 7.604562737642586e-05, |
| "loss": 0.8782, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.47573739295908657, |
| "grad_norm": 193025.796875, |
| "learning_rate": 9.505703422053233e-05, |
| "loss": 0.7562, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.570884871550904, |
| "grad_norm": 136630.859375, |
| "learning_rate": 9.843518714315923e-05, |
| "loss": 0.603, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.6660323501427212, |
| "grad_norm": 175707.984375, |
| "learning_rate": 9.63205751744555e-05, |
| "loss": 0.5252, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.7611798287345385, |
| "grad_norm": 199550.734375, |
| "learning_rate": 9.420596320575174e-05, |
| "loss": 0.4753, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.8563273073263559, |
| "grad_norm": 147047.765625, |
| "learning_rate": 9.2091351237048e-05, |
| "loss": 0.4839, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.9514747859181731, |
| "grad_norm": 170235.53125, |
| "learning_rate": 8.997673926834426e-05, |
| "loss": 0.4608, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8310730430644777, |
| "eval_f1_macro": 0.8119798637764705, |
| "eval_loss": 0.397123783826828, |
| "eval_runtime": 29.8024, |
| "eval_samples_per_second": 141.029, |
| "eval_steps_per_second": 4.429, |
| "step": 1051 |
| }, |
| { |
| "epoch": 1.0466222645099905, |
| "grad_norm": 154165.984375, |
| "learning_rate": 8.786212729964052e-05, |
| "loss": 0.4329, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.141769743101808, |
| "grad_norm": 167387.453125, |
| "learning_rate": 8.574751533093678e-05, |
| "loss": 0.4237, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.236917221693625, |
| "grad_norm": 78232.890625, |
| "learning_rate": 8.363290336223304e-05, |
| "loss": 0.4274, |
| "step": 1300 |
| }, |
| { |
| "epoch": 1.3320647002854424, |
| "grad_norm": 157848.515625, |
| "learning_rate": 8.151829139352929e-05, |
| "loss": 0.4101, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.4272121788772598, |
| "grad_norm": 83062.8359375, |
| "learning_rate": 7.940367942482555e-05, |
| "loss": 0.4046, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.5223596574690772, |
| "grad_norm": 217329.375, |
| "learning_rate": 7.728906745612181e-05, |
| "loss": 0.405, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.6175071360608944, |
| "grad_norm": 219439.40625, |
| "learning_rate": 7.517445548741806e-05, |
| "loss": 0.3994, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.7126546146527117, |
| "grad_norm": 94135.8046875, |
| "learning_rate": 7.305984351871432e-05, |
| "loss": 0.3957, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.8078020932445291, |
| "grad_norm": 267690.875, |
| "learning_rate": 7.094523155001058e-05, |
| "loss": 0.4029, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.9029495718363463, |
| "grad_norm": 289756.53125, |
| "learning_rate": 6.883061958130684e-05, |
| "loss": 0.3973, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.9980970504281637, |
| "grad_norm": 230541.8125, |
| "learning_rate": 6.67160076126031e-05, |
| "loss": 0.3772, |
| "step": 2100 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8505829169640733, |
| "eval_f1_macro": 0.8312500853233197, |
| "eval_loss": 0.3541408181190491, |
| "eval_runtime": 30.0593, |
| "eval_samples_per_second": 139.824, |
| "eval_steps_per_second": 4.391, |
| "step": 2102 |
| }, |
| { |
| "epoch": 2.093244529019981, |
| "grad_norm": 163747.640625, |
| "learning_rate": 6.460139564389934e-05, |
| "loss": 0.3638, |
| "step": 2200 |
| }, |
| { |
| "epoch": 2.188392007611798, |
| "grad_norm": 269249.65625, |
| "learning_rate": 6.24867836751956e-05, |
| "loss": 0.367, |
| "step": 2300 |
| }, |
| { |
| "epoch": 2.283539486203616, |
| "grad_norm": 131646.234375, |
| "learning_rate": 6.0372171706491863e-05, |
| "loss": 0.3669, |
| "step": 2400 |
| }, |
| { |
| "epoch": 2.378686964795433, |
| "grad_norm": 163041.328125, |
| "learning_rate": 5.825755973778812e-05, |
| "loss": 0.3556, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.47383444338725, |
| "grad_norm": 183448.84375, |
| "learning_rate": 5.614294776908438e-05, |
| "loss": 0.3565, |
| "step": 2600 |
| }, |
| { |
| "epoch": 2.5689819219790677, |
| "grad_norm": 223062.75, |
| "learning_rate": 5.402833580038064e-05, |
| "loss": 0.3728, |
| "step": 2700 |
| }, |
| { |
| "epoch": 2.664129400570885, |
| "grad_norm": 125715.6640625, |
| "learning_rate": 5.1913723831676884e-05, |
| "loss": 0.3654, |
| "step": 2800 |
| }, |
| { |
| "epoch": 2.759276879162702, |
| "grad_norm": 179914.296875, |
| "learning_rate": 4.9799111862973144e-05, |
| "loss": 0.3571, |
| "step": 2900 |
| }, |
| { |
| "epoch": 2.8544243577545196, |
| "grad_norm": 184159.671875, |
| "learning_rate": 4.7684499894269404e-05, |
| "loss": 0.3499, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.949571836346337, |
| "grad_norm": 123080.375, |
| "learning_rate": 4.5569887925565665e-05, |
| "loss": 0.3556, |
| "step": 3100 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8624791815369974, |
| "eval_f1_macro": 0.8367348377738546, |
| "eval_loss": 0.32585257291793823, |
| "eval_runtime": 29.8922, |
| "eval_samples_per_second": 140.605, |
| "eval_steps_per_second": 4.416, |
| "step": 3153 |
| }, |
| { |
| "epoch": 3.044719314938154, |
| "grad_norm": 232850.921875, |
| "learning_rate": 4.345527595686192e-05, |
| "loss": 0.3615, |
| "step": 3200 |
| }, |
| { |
| "epoch": 3.1398667935299716, |
| "grad_norm": 123978.3515625, |
| "learning_rate": 4.134066398815817e-05, |
| "loss": 0.3419, |
| "step": 3300 |
| }, |
| { |
| "epoch": 3.2350142721217887, |
| "grad_norm": 143989.796875, |
| "learning_rate": 3.922605201945443e-05, |
| "loss": 0.3339, |
| "step": 3400 |
| }, |
| { |
| "epoch": 3.3301617507136063, |
| "grad_norm": 131409.40625, |
| "learning_rate": 3.711144005075069e-05, |
| "loss": 0.3272, |
| "step": 3500 |
| }, |
| { |
| "epoch": 3.4253092293054235, |
| "grad_norm": 295399.4375, |
| "learning_rate": 3.4996828082046945e-05, |
| "loss": 0.3529, |
| "step": 3600 |
| }, |
| { |
| "epoch": 3.5204567078972406, |
| "grad_norm": 279721.375, |
| "learning_rate": 3.28822161133432e-05, |
| "loss": 0.3484, |
| "step": 3700 |
| }, |
| { |
| "epoch": 3.6156041864890582, |
| "grad_norm": 234246.640625, |
| "learning_rate": 3.076760414463946e-05, |
| "loss": 0.3274, |
| "step": 3800 |
| }, |
| { |
| "epoch": 3.7107516650808754, |
| "grad_norm": 215512.46875, |
| "learning_rate": 2.8652992175935716e-05, |
| "loss": 0.3289, |
| "step": 3900 |
| }, |
| { |
| "epoch": 3.8058991436726926, |
| "grad_norm": 193919.359375, |
| "learning_rate": 2.6538380207231972e-05, |
| "loss": 0.3427, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.90104662226451, |
| "grad_norm": 266348.53125, |
| "learning_rate": 2.442376823852823e-05, |
| "loss": 0.3488, |
| "step": 4100 |
| }, |
| { |
| "epoch": 3.9961941008563273, |
| "grad_norm": 157798.1875, |
| "learning_rate": 2.230915626982449e-05, |
| "loss": 0.327, |
| "step": 4200 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8639067332857483, |
| "eval_f1_macro": 0.8404976856282999, |
| "eval_loss": 0.3265078663825989, |
| "eval_runtime": 29.6057, |
| "eval_samples_per_second": 141.966, |
| "eval_steps_per_second": 4.459, |
| "step": 4204 |
| }, |
| { |
| "epoch": 4.0913415794481445, |
| "grad_norm": 93288.796875, |
| "learning_rate": 2.0194544301120746e-05, |
| "loss": 0.3462, |
| "step": 4300 |
| }, |
| { |
| "epoch": 4.186489058039962, |
| "grad_norm": 155794.875, |
| "learning_rate": 1.8079932332417003e-05, |
| "loss": 0.3407, |
| "step": 4400 |
| }, |
| { |
| "epoch": 4.28163653663178, |
| "grad_norm": 180881.90625, |
| "learning_rate": 1.596532036371326e-05, |
| "loss": 0.3082, |
| "step": 4500 |
| }, |
| { |
| "epoch": 4.376784015223596, |
| "grad_norm": 150323.046875, |
| "learning_rate": 1.3850708395009515e-05, |
| "loss": 0.315, |
| "step": 4600 |
| }, |
| { |
| "epoch": 4.471931493815414, |
| "grad_norm": 123327.1796875, |
| "learning_rate": 1.1736096426305774e-05, |
| "loss": 0.3391, |
| "step": 4700 |
| }, |
| { |
| "epoch": 4.567078972407232, |
| "grad_norm": 160687.984375, |
| "learning_rate": 9.62148445760203e-06, |
| "loss": 0.3163, |
| "step": 4800 |
| }, |
| { |
| "epoch": 4.662226450999048, |
| "grad_norm": 59321.7109375, |
| "learning_rate": 7.506872488898288e-06, |
| "loss": 0.3198, |
| "step": 4900 |
| }, |
| { |
| "epoch": 4.757373929590866, |
| "grad_norm": 117285.140625, |
| "learning_rate": 5.392260520194545e-06, |
| "loss": 0.3315, |
| "step": 5000 |
| }, |
| { |
| "epoch": 4.8525214081826835, |
| "grad_norm": 153029.640625, |
| "learning_rate": 3.2776485514908016e-06, |
| "loss": 0.3144, |
| "step": 5100 |
| }, |
| { |
| "epoch": 4.9476688867745, |
| "grad_norm": 82521.359375, |
| "learning_rate": 1.1630365827870586e-06, |
| "loss": 0.3221, |
| "step": 5200 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8655722103259577, |
| "eval_f1_macro": 0.8411684097031022, |
| "eval_loss": 0.319923996925354, |
| "eval_runtime": 29.7497, |
| "eval_samples_per_second": 141.279, |
| "eval_steps_per_second": 4.437, |
| "step": 5255 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 5255, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 2, |
| "early_stopping_threshold": 0.0 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 0 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.125063421341696e+16, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
