{ "best_metric": 0.7240942716598511, "best_model_checkpoint": "v17/checkpoint-264", "epoch": 2.0, "eval_steps": 500, "global_step": 264, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.045454545454545456, "grad_norm": Infinity, "learning_rate": 7.575757575757576e-07, "loss": 3.3291, "step": 6 }, { "epoch": 0.09090909090909091, "grad_norm": 102.48841094970703, "learning_rate": 2.0202020202020206e-06, "loss": 2.9882, "step": 12 }, { "epoch": 0.13636363636363635, "grad_norm": 43.2066764831543, "learning_rate": 3.5353535353535352e-06, "loss": 3.0503, "step": 18 }, { "epoch": 0.18181818181818182, "grad_norm": 30.021760940551758, "learning_rate": 5.050505050505051e-06, "loss": 2.3939, "step": 24 }, { "epoch": 0.22727272727272727, "grad_norm": 37.78391647338867, "learning_rate": 6.565656565656567e-06, "loss": 2.4729, "step": 30 }, { "epoch": 0.2727272727272727, "grad_norm": 69.5600357055664, "learning_rate": 8.080808080808082e-06, "loss": 1.5488, "step": 36 }, { "epoch": 0.3181818181818182, "grad_norm": 35.23645782470703, "learning_rate": 9.595959595959595e-06, "loss": 1.2906, "step": 42 }, { "epoch": 0.36363636363636365, "grad_norm": 39.70997619628906, "learning_rate": 1.1111111111111112e-05, "loss": 1.3436, "step": 48 }, { "epoch": 0.4090909090909091, "grad_norm": 14.217961311340332, "learning_rate": 1.2626262626262628e-05, "loss": 0.7839, "step": 54 }, { "epoch": 0.45454545454545453, "grad_norm": 22.282686233520508, "learning_rate": 1.4141414141414141e-05, "loss": 0.95, "step": 60 }, { "epoch": 0.5, "grad_norm": 32.02109146118164, "learning_rate": 1.565656565656566e-05, "loss": 0.939, "step": 66 }, { "epoch": 0.5454545454545454, "grad_norm": 14.312190055847168, "learning_rate": 1.7171717171717173e-05, "loss": 0.7151, "step": 72 }, { "epoch": 0.5909090909090909, "grad_norm": 11.623531341552734, "learning_rate": 1.8686868686868688e-05, "loss": 0.8836, "step": 78 }, { "epoch": 0.6363636363636364, "grad_norm": 5.483341217041016, "learning_rate": 2.0202020202020203e-05, "loss": 0.8209, "step": 84 }, { "epoch": 0.6818181818181818, "grad_norm": 13.19078540802002, "learning_rate": 2.171717171717172e-05, "loss": 0.6938, "step": 90 }, { "epoch": 0.7272727272727273, "grad_norm": 17.33362579345703, "learning_rate": 2.3232323232323232e-05, "loss": 0.8035, "step": 96 }, { "epoch": 0.7727272727272727, "grad_norm": 21.31452751159668, "learning_rate": 2.474747474747475e-05, "loss": 0.7983, "step": 102 }, { "epoch": 0.8181818181818182, "grad_norm": 28.60481834411621, "learning_rate": 2.6262626262626268e-05, "loss": 0.7791, "step": 108 }, { "epoch": 0.8636363636363636, "grad_norm": 5.379669666290283, "learning_rate": 2.777777777777778e-05, "loss": 0.5558, "step": 114 }, { "epoch": 0.9090909090909091, "grad_norm": 14.164498329162598, "learning_rate": 2.9292929292929294e-05, "loss": 0.6367, "step": 120 }, { "epoch": 0.9545454545454546, "grad_norm": 19.893098831176758, "learning_rate": 3.080808080808081e-05, "loss": 0.728, "step": 126 }, { "epoch": 1.0, "grad_norm": 18.595251083374023, "learning_rate": 3.232323232323233e-05, "loss": 0.6763, "step": 132 }, { "epoch": 1.0, "eval_accuracy": 0.6222222222222222, "eval_f1_macro": 0.5643078521152306, "eval_f1_micro": 0.6222222222222222, "eval_f1_weighted": 0.6277790773440842, "eval_loss": 0.8971771597862244, "eval_precision_macro": 0.5661957090528519, "eval_precision_micro": 0.6222222222222222, "eval_precision_weighted": 0.7270577746768223, "eval_recall_macro": 0.7581088510170423, "eval_recall_micro": 0.6222222222222222, "eval_recall_weighted": 0.6222222222222222, "eval_runtime": 0.7127, "eval_samples_per_second": 378.823, "eval_steps_per_second": 23.852, "step": 132 }, { "epoch": 1.0454545454545454, "grad_norm": 9.349661827087402, "learning_rate": 3.3838383838383844e-05, "loss": 0.7284, "step": 138 }, { "epoch": 1.0909090909090908, "grad_norm": 7.680129051208496, "learning_rate": 3.535353535353535e-05, "loss": 0.4757, "step": 144 }, { "epoch": 1.1363636363636362, "grad_norm": 5.932518005371094, "learning_rate": 3.686868686868687e-05, "loss": 0.5222, "step": 150 }, { "epoch": 1.1818181818181819, "grad_norm": 23.951522827148438, "learning_rate": 3.838383838383838e-05, "loss": 0.5293, "step": 156 }, { "epoch": 1.2272727272727273, "grad_norm": 6.550896167755127, "learning_rate": 3.98989898989899e-05, "loss": 0.6098, "step": 162 }, { "epoch": 1.2727272727272727, "grad_norm": 8.644554138183594, "learning_rate": 4.141414141414142e-05, "loss": 0.509, "step": 168 }, { "epoch": 1.3181818181818181, "grad_norm": 35.59321594238281, "learning_rate": 4.292929292929293e-05, "loss": 0.7367, "step": 174 }, { "epoch": 1.3636363636363638, "grad_norm": 11.410784721374512, "learning_rate": 4.4444444444444447e-05, "loss": 0.6406, "step": 180 }, { "epoch": 1.4090909090909092, "grad_norm": 18.567852020263672, "learning_rate": 4.595959595959596e-05, "loss": 0.6903, "step": 186 }, { "epoch": 1.4545454545454546, "grad_norm": 5.068539619445801, "learning_rate": 4.7474747474747476e-05, "loss": 0.5961, "step": 192 }, { "epoch": 1.5, "grad_norm": 8.43880844116211, "learning_rate": 4.898989898989899e-05, "loss": 0.5387, "step": 198 }, { "epoch": 1.5454545454545454, "grad_norm": 10.733231544494629, "learning_rate": 4.994388327721661e-05, "loss": 0.6114, "step": 204 }, { "epoch": 1.5909090909090908, "grad_norm": 13.800358772277832, "learning_rate": 4.977553310886644e-05, "loss": 0.6206, "step": 210 }, { "epoch": 1.6363636363636362, "grad_norm": 11.480915069580078, "learning_rate": 4.960718294051628e-05, "loss": 0.6728, "step": 216 }, { "epoch": 1.6818181818181817, "grad_norm": 8.072641372680664, "learning_rate": 4.943883277216611e-05, "loss": 0.501, "step": 222 }, { "epoch": 1.7272727272727273, "grad_norm": 20.036832809448242, "learning_rate": 4.927048260381594e-05, "loss": 0.4993, "step": 228 }, { "epoch": 1.7727272727272727, "grad_norm": 3.0253636837005615, "learning_rate": 4.910213243546577e-05, "loss": 0.6286, "step": 234 }, { "epoch": 1.8181818181818183, "grad_norm": 11.858734130859375, "learning_rate": 4.89337822671156e-05, "loss": 0.5855, "step": 240 }, { "epoch": 1.8636363636363638, "grad_norm": 18.448442459106445, "learning_rate": 4.876543209876544e-05, "loss": 0.7015, "step": 246 }, { "epoch": 1.9090909090909092, "grad_norm": 12.252097129821777, "learning_rate": 4.859708193041527e-05, "loss": 0.7009, "step": 252 }, { "epoch": 1.9545454545454546, "grad_norm": 8.62966251373291, "learning_rate": 4.84287317620651e-05, "loss": 0.7572, "step": 258 }, { "epoch": 2.0, "grad_norm": 15.64622974395752, "learning_rate": 4.8260381593714935e-05, "loss": 0.7993, "step": 264 }, { "epoch": 2.0, "eval_accuracy": 0.5888888888888889, "eval_f1_macro": 0.6317920646698781, "eval_f1_micro": 0.5888888888888889, "eval_f1_weighted": 0.5609562339331899, "eval_loss": 0.7240942716598511, "eval_precision_macro": 0.714913844325609, "eval_precision_micro": 0.5888888888888889, "eval_precision_weighted": 0.7466534627318941, "eval_recall_macro": 0.6893042575285566, "eval_recall_micro": 0.5888888888888889, "eval_recall_weighted": 0.5888888888888889, "eval_runtime": 0.7078, "eval_samples_per_second": 381.47, "eval_steps_per_second": 24.019, "step": 264 } ], "logging_steps": 6, "max_steps": 1980, "num_input_tokens_seen": 0, "num_train_epochs": 15, "save_steps": 500, "total_flos": 138134544307200.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }