{ "best_metric": 0.587009608745575, "best_model_checkpoint": "V13/checkpoint-312", "epoch": 2.0, "eval_steps": 500, "global_step": 312, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.04487179487179487, "grad_norm": 63.9647331237793, "learning_rate": 6.41025641025641e-07, "loss": 3.0773, "step": 7 }, { "epoch": 0.08974358974358974, "grad_norm": 58.97015380859375, "learning_rate": 2.136752136752137e-06, "loss": 2.7492, "step": 14 }, { "epoch": 0.1346153846153846, "grad_norm": 64.63655090332031, "learning_rate": 3.632478632478633e-06, "loss": 2.1906, "step": 21 }, { "epoch": 0.1794871794871795, "grad_norm": 29.906232833862305, "learning_rate": 5.128205128205128e-06, "loss": 2.3421, "step": 28 }, { "epoch": 0.22435897435897437, "grad_norm": 22.833885192871094, "learning_rate": 6.623931623931625e-06, "loss": 1.7169, "step": 35 }, { "epoch": 0.2692307692307692, "grad_norm": 30.393211364746094, "learning_rate": 8.11965811965812e-06, "loss": 1.6294, "step": 42 }, { "epoch": 0.3141025641025641, "grad_norm": 38.7406005859375, "learning_rate": 9.615384615384616e-06, "loss": 1.6305, "step": 49 }, { "epoch": 0.358974358974359, "grad_norm": 13.8955659866333, "learning_rate": 1.1111111111111112e-05, "loss": 1.3634, "step": 56 }, { "epoch": 0.40384615384615385, "grad_norm": 19.50594711303711, "learning_rate": 1.2606837606837608e-05, "loss": 1.0247, "step": 63 }, { "epoch": 0.44871794871794873, "grad_norm": 9.244388580322266, "learning_rate": 1.4102564102564104e-05, "loss": 0.752, "step": 70 }, { "epoch": 0.4935897435897436, "grad_norm": 9.010939598083496, "learning_rate": 1.5598290598290598e-05, "loss": 0.6144, "step": 77 }, { "epoch": 0.5384615384615384, "grad_norm": 16.850486755371094, "learning_rate": 1.7094017094017095e-05, "loss": 0.8148, "step": 84 }, { "epoch": 0.5833333333333334, "grad_norm": 14.541914939880371, "learning_rate": 1.858974358974359e-05, "loss": 0.7843, "step": 91 }, { "epoch": 0.6282051282051282, "grad_norm": 15.613171577453613, "learning_rate": 2.0085470085470087e-05, "loss": 0.7653, "step": 98 }, { "epoch": 0.6730769230769231, "grad_norm": 17.819730758666992, "learning_rate": 2.1581196581196585e-05, "loss": 0.8886, "step": 105 }, { "epoch": 0.717948717948718, "grad_norm": 25.498430252075195, "learning_rate": 2.307692307692308e-05, "loss": 0.8585, "step": 112 }, { "epoch": 0.7628205128205128, "grad_norm": 6.939703464508057, "learning_rate": 2.4572649572649573e-05, "loss": 0.7819, "step": 119 }, { "epoch": 0.8076923076923077, "grad_norm": 5.337686061859131, "learning_rate": 2.606837606837607e-05, "loss": 0.7656, "step": 126 }, { "epoch": 0.8525641025641025, "grad_norm": 5.561918258666992, "learning_rate": 2.756410256410257e-05, "loss": 0.6922, "step": 133 }, { "epoch": 0.8974358974358975, "grad_norm": 4.981381416320801, "learning_rate": 2.9059829059829063e-05, "loss": 0.6088, "step": 140 }, { "epoch": 0.9423076923076923, "grad_norm": 10.54244613647461, "learning_rate": 3.055555555555556e-05, "loss": 0.6644, "step": 147 }, { "epoch": 0.9871794871794872, "grad_norm": 18.519126892089844, "learning_rate": 3.205128205128206e-05, "loss": 0.6198, "step": 154 }, { "epoch": 1.0, "eval_accuracy": 0.6955128205128205, "eval_f1_macro": 0.6747979178342244, "eval_f1_micro": 0.6955128205128205, "eval_f1_weighted": 0.6888607024416085, "eval_loss": 0.6790350675582886, "eval_precision_macro": 0.79043461935605, "eval_precision_micro": 0.6955128205128205, "eval_precision_weighted": 0.6978595772949476, "eval_recall_macro": 0.6219737626714371, "eval_recall_micro": 0.6955128205128205, "eval_recall_weighted": 0.6955128205128205, "eval_runtime": 0.8327, "eval_samples_per_second": 374.664, "eval_steps_per_second": 24.017, "step": 156 }, { "epoch": 1.032051282051282, "grad_norm": 17.723602294921875, "learning_rate": 3.354700854700855e-05, "loss": 0.7364, "step": 161 }, { "epoch": 1.0769230769230769, "grad_norm": 54.3127326965332, "learning_rate": 3.504273504273504e-05, "loss": 0.6054, "step": 168 }, { "epoch": 1.1217948717948718, "grad_norm": 7.926159381866455, "learning_rate": 3.653846153846154e-05, "loss": 0.54, "step": 175 }, { "epoch": 1.1666666666666667, "grad_norm": 6.779534816741943, "learning_rate": 3.8034188034188035e-05, "loss": 0.7181, "step": 182 }, { "epoch": 1.2115384615384615, "grad_norm": 4.455146789550781, "learning_rate": 3.952991452991453e-05, "loss": 0.57, "step": 189 }, { "epoch": 1.2564102564102564, "grad_norm": 11.916348457336426, "learning_rate": 4.1025641025641023e-05, "loss": 0.6322, "step": 196 }, { "epoch": 1.3012820512820513, "grad_norm": 11.289109230041504, "learning_rate": 4.252136752136752e-05, "loss": 0.5207, "step": 203 }, { "epoch": 1.3461538461538463, "grad_norm": 9.974486351013184, "learning_rate": 4.401709401709402e-05, "loss": 0.6792, "step": 210 }, { "epoch": 1.391025641025641, "grad_norm": 7.0826640129089355, "learning_rate": 4.5512820512820516e-05, "loss": 0.6701, "step": 217 }, { "epoch": 1.435897435897436, "grad_norm": 16.46434783935547, "learning_rate": 4.700854700854701e-05, "loss": 0.5722, "step": 224 }, { "epoch": 1.4807692307692308, "grad_norm": 12.298468589782715, "learning_rate": 4.8504273504273505e-05, "loss": 0.7939, "step": 231 }, { "epoch": 1.5256410256410255, "grad_norm": 25.05532455444336, "learning_rate": 5e-05, "loss": 0.7046, "step": 238 }, { "epoch": 1.5705128205128205, "grad_norm": 31.899816513061523, "learning_rate": 4.983380816714151e-05, "loss": 0.5815, "step": 245 }, { "epoch": 1.6153846153846154, "grad_norm": 6.716318607330322, "learning_rate": 4.9667616334283e-05, "loss": 0.5213, "step": 252 }, { "epoch": 1.6602564102564101, "grad_norm": 8.302348136901855, "learning_rate": 4.95014245014245e-05, "loss": 0.5883, "step": 259 }, { "epoch": 1.7051282051282053, "grad_norm": 40.705841064453125, "learning_rate": 4.933523266856601e-05, "loss": 0.5492, "step": 266 }, { "epoch": 1.75, "grad_norm": 7.286283493041992, "learning_rate": 4.9169040835707506e-05, "loss": 0.6444, "step": 273 }, { "epoch": 1.7948717948717947, "grad_norm": 9.41189956665039, "learning_rate": 4.9002849002849004e-05, "loss": 0.6681, "step": 280 }, { "epoch": 1.8397435897435899, "grad_norm": 12.600714683532715, "learning_rate": 4.883665716999051e-05, "loss": 0.6835, "step": 287 }, { "epoch": 1.8846153846153846, "grad_norm": 6.682600498199463, "learning_rate": 4.867046533713201e-05, "loss": 0.7121, "step": 294 }, { "epoch": 1.9294871794871795, "grad_norm": 15.127613067626953, "learning_rate": 4.8504273504273505e-05, "loss": 0.5505, "step": 301 }, { "epoch": 1.9743589743589745, "grad_norm": 9.874059677124023, "learning_rate": 4.833808167141501e-05, "loss": 0.5549, "step": 308 }, { "epoch": 2.0, "eval_accuracy": 0.7435897435897436, "eval_f1_macro": 0.7086614173228346, "eval_f1_micro": 0.7435897435897436, "eval_f1_weighted": 0.7361195235210982, "eval_loss": 0.587009608745575, "eval_precision_macro": 0.7210665002427016, "eval_precision_micro": 0.7435897435897436, "eval_precision_weighted": 0.7538358857008742, "eval_recall_macro": 0.7157125819916518, "eval_recall_micro": 0.7435897435897436, "eval_recall_weighted": 0.7435897435897436, "eval_runtime": 0.8226, "eval_samples_per_second": 379.293, "eval_steps_per_second": 24.314, "step": 312 } ], "logging_steps": 7, "max_steps": 2340, "num_input_tokens_seen": 0, "num_train_epochs": 15, "save_steps": 500, "total_flos": 164182772662272.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }