| { | |
| "best_metric": 0.7293202496323148, | |
| "best_model_checkpoint": "training_dir/checkpoint-16000", | |
| "epoch": 2.0133517007523576, | |
| "eval_steps": 1000, | |
| "global_step": 19000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1059658789869662, | |
| "grad_norm": 15.513019561767578, | |
| "learning_rate": 9.946734846063707e-06, | |
| "loss": 0.8105, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1059658789869662, | |
| "eval_accuracy": 0.6267482517482518, | |
| "eval_f1": 0.5486997676789124, | |
| "eval_loss": 0.8523119688034058, | |
| "eval_precision": 0.6663256334906192, | |
| "eval_recall": 0.6267482517482518, | |
| "eval_runtime": 26.6488, | |
| "eval_samples_per_second": 85.857, | |
| "eval_steps_per_second": 10.732, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2119317579739324, | |
| "grad_norm": 14.981549263000488, | |
| "learning_rate": 9.840204538191116e-06, | |
| "loss": 0.5828, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2119317579739324, | |
| "eval_accuracy": 0.6682692307692307, | |
| "eval_f1": 0.6366336108184594, | |
| "eval_loss": 0.9371851086616516, | |
| "eval_precision": 0.6852715704153842, | |
| "eval_recall": 0.6682692307692307, | |
| "eval_runtime": 25.9805, | |
| "eval_samples_per_second": 88.066, | |
| "eval_steps_per_second": 11.008, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.31789763696089857, | |
| "grad_norm": 8.239383697509766, | |
| "learning_rate": 9.733674230318526e-06, | |
| "loss": 0.5349, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.31789763696089857, | |
| "eval_accuracy": 0.6914335664335665, | |
| "eval_f1": 0.669233772837662, | |
| "eval_loss": 0.8858013153076172, | |
| "eval_precision": 0.6967032053774095, | |
| "eval_recall": 0.6914335664335665, | |
| "eval_runtime": 25.9486, | |
| "eval_samples_per_second": 88.174, | |
| "eval_steps_per_second": 11.022, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.4238635159478648, | |
| "grad_norm": 9.338201522827148, | |
| "learning_rate": 9.627143922445936e-06, | |
| "loss": 0.5233, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4238635159478648, | |
| "eval_accuracy": 0.7023601398601399, | |
| "eval_f1": 0.6877643727244785, | |
| "eval_loss": 0.9143859148025513, | |
| "eval_precision": 0.7132978004907108, | |
| "eval_recall": 0.7023601398601399, | |
| "eval_runtime": 26.1039, | |
| "eval_samples_per_second": 87.65, | |
| "eval_steps_per_second": 10.956, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.5298293949348309, | |
| "grad_norm": 7.548835754394531, | |
| "learning_rate": 9.520613614573348e-06, | |
| "loss": 0.5089, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5298293949348309, | |
| "eval_accuracy": 0.7084790209790209, | |
| "eval_f1": 0.6979166210878224, | |
| "eval_loss": 0.8279535174369812, | |
| "eval_precision": 0.7081949635182484, | |
| "eval_recall": 0.7084790209790209, | |
| "eval_runtime": 25.9523, | |
| "eval_samples_per_second": 88.162, | |
| "eval_steps_per_second": 11.02, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.6357952739217971, | |
| "grad_norm": 7.738587379455566, | |
| "learning_rate": 9.414083306700758e-06, | |
| "loss": 0.4964, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6357952739217971, | |
| "eval_accuracy": 0.7106643356643356, | |
| "eval_f1": 0.6933251320034988, | |
| "eval_loss": 0.7791977524757385, | |
| "eval_precision": 0.7149995516226468, | |
| "eval_recall": 0.7106643356643356, | |
| "eval_runtime": 26.0573, | |
| "eval_samples_per_second": 87.806, | |
| "eval_steps_per_second": 10.976, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.7417611529087634, | |
| "grad_norm": 2.134586811065674, | |
| "learning_rate": 9.307552998828167e-06, | |
| "loss": 0.4972, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7417611529087634, | |
| "eval_accuracy": 0.715472027972028, | |
| "eval_f1": 0.7023119335505582, | |
| "eval_loss": 0.8132364749908447, | |
| "eval_precision": 0.7209094583908723, | |
| "eval_recall": 0.715472027972028, | |
| "eval_runtime": 26.036, | |
| "eval_samples_per_second": 87.878, | |
| "eval_steps_per_second": 10.985, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.8477270318957296, | |
| "grad_norm": 7.15505838394165, | |
| "learning_rate": 9.201022690955577e-06, | |
| "loss": 0.4598, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.8477270318957296, | |
| "eval_accuracy": 0.7263986013986014, | |
| "eval_f1": 0.7183207652218417, | |
| "eval_loss": 0.782526969909668, | |
| "eval_precision": 0.7270872567100993, | |
| "eval_recall": 0.7263986013986014, | |
| "eval_runtime": 26.0044, | |
| "eval_samples_per_second": 87.985, | |
| "eval_steps_per_second": 10.998, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.9536929108826958, | |
| "grad_norm": 4.955699920654297, | |
| "learning_rate": 9.094492383082987e-06, | |
| "loss": 0.4648, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.9536929108826958, | |
| "eval_accuracy": 0.7167832167832168, | |
| "eval_f1": 0.7085336559753441, | |
| "eval_loss": 0.8246528506278992, | |
| "eval_precision": 0.7230635739304796, | |
| "eval_recall": 0.7167832167832168, | |
| "eval_runtime": 26.0039, | |
| "eval_samples_per_second": 87.987, | |
| "eval_steps_per_second": 10.998, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.0596587898696619, | |
| "grad_norm": 10.622162818908691, | |
| "learning_rate": 8.987962075210399e-06, | |
| "loss": 0.4261, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0596587898696619, | |
| "eval_accuracy": 0.7242132867132867, | |
| "eval_f1": 0.7187402398550484, | |
| "eval_loss": 0.8997145295143127, | |
| "eval_precision": 0.7327413713313974, | |
| "eval_recall": 0.7242132867132867, | |
| "eval_runtime": 26.0449, | |
| "eval_samples_per_second": 87.848, | |
| "eval_steps_per_second": 10.981, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.165624668856628, | |
| "grad_norm": 5.598655700683594, | |
| "learning_rate": 8.881431767337809e-06, | |
| "loss": 0.3778, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.165624668856628, | |
| "eval_accuracy": 0.7202797202797203, | |
| "eval_f1": 0.7064609677451916, | |
| "eval_loss": 1.072676420211792, | |
| "eval_precision": 0.7267989053042323, | |
| "eval_recall": 0.7202797202797203, | |
| "eval_runtime": 26.0924, | |
| "eval_samples_per_second": 87.688, | |
| "eval_steps_per_second": 10.961, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.2715905478435943, | |
| "grad_norm": 3.347665548324585, | |
| "learning_rate": 8.774901459465218e-06, | |
| "loss": 0.4096, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.2715905478435943, | |
| "eval_accuracy": 0.7338286713286714, | |
| "eval_f1": 0.7260786855980778, | |
| "eval_loss": 0.8679103851318359, | |
| "eval_precision": 0.7338714881942328, | |
| "eval_recall": 0.7338286713286714, | |
| "eval_runtime": 26.1513, | |
| "eval_samples_per_second": 87.491, | |
| "eval_steps_per_second": 10.936, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.3775564268305605, | |
| "grad_norm": 15.350415229797363, | |
| "learning_rate": 8.668371151592628e-06, | |
| "loss": 0.3868, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.3775564268305605, | |
| "eval_accuracy": 0.7268356643356644, | |
| "eval_f1": 0.7121431239588121, | |
| "eval_loss": 0.9578595161437988, | |
| "eval_precision": 0.7318944469113846, | |
| "eval_recall": 0.7268356643356644, | |
| "eval_runtime": 25.9641, | |
| "eval_samples_per_second": 88.122, | |
| "eval_steps_per_second": 11.015, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.4835223058175266, | |
| "grad_norm": 19.44073486328125, | |
| "learning_rate": 8.56184084372004e-06, | |
| "loss": 0.3967, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.4835223058175266, | |
| "eval_accuracy": 0.7377622377622378, | |
| "eval_f1": 0.7275954782820151, | |
| "eval_loss": 0.9781323671340942, | |
| "eval_precision": 0.7423521044319988, | |
| "eval_recall": 0.7377622377622378, | |
| "eval_runtime": 29.6474, | |
| "eval_samples_per_second": 77.174, | |
| "eval_steps_per_second": 9.647, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.5894881848044928, | |
| "grad_norm": 0.27897197008132935, | |
| "learning_rate": 8.45531053584745e-06, | |
| "loss": 0.4048, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.5894881848044928, | |
| "eval_accuracy": 0.7246503496503497, | |
| "eval_f1": 0.7134343410409786, | |
| "eval_loss": 1.0050002336502075, | |
| "eval_precision": 0.7324782904105448, | |
| "eval_recall": 0.7246503496503497, | |
| "eval_runtime": 28.5313, | |
| "eval_samples_per_second": 80.192, | |
| "eval_steps_per_second": 10.024, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.695454063791459, | |
| "grad_norm": 17.813819885253906, | |
| "learning_rate": 8.34878022797486e-06, | |
| "loss": 0.4002, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.695454063791459, | |
| "eval_accuracy": 0.7351398601398601, | |
| "eval_f1": 0.7293202496323148, | |
| "eval_loss": 0.9046709537506104, | |
| "eval_precision": 0.7368475219229782, | |
| "eval_recall": 0.7351398601398601, | |
| "eval_runtime": 29.0485, | |
| "eval_samples_per_second": 78.765, | |
| "eval_steps_per_second": 9.846, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.8014199427784252, | |
| "grad_norm": 11.392443656921387, | |
| "learning_rate": 8.24224992010227e-06, | |
| "loss": 0.3906, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.8014199427784252, | |
| "eval_accuracy": 0.7312062937062938, | |
| "eval_f1": 0.7210893936836806, | |
| "eval_loss": 0.9010310769081116, | |
| "eval_precision": 0.7345691438075613, | |
| "eval_recall": 0.7312062937062938, | |
| "eval_runtime": 28.5861, | |
| "eval_samples_per_second": 80.039, | |
| "eval_steps_per_second": 10.005, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.9073858217653914, | |
| "grad_norm": 10.869200706481934, | |
| "learning_rate": 8.13571961222968e-06, | |
| "loss": 0.3847, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.9073858217653914, | |
| "eval_accuracy": 0.7229020979020979, | |
| "eval_f1": 0.7136781919341553, | |
| "eval_loss": 0.9686246514320374, | |
| "eval_precision": 0.7275042194736051, | |
| "eval_recall": 0.7229020979020979, | |
| "eval_runtime": 28.6996, | |
| "eval_samples_per_second": 79.722, | |
| "eval_steps_per_second": 9.965, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.0133517007523576, | |
| "grad_norm": 12.705787658691406, | |
| "learning_rate": 8.02918930435709e-06, | |
| "loss": 0.3744, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.0133517007523576, | |
| "eval_accuracy": 0.7259615384615384, | |
| "eval_f1": 0.7157971512813134, | |
| "eval_loss": 1.0338144302368164, | |
| "eval_precision": 0.7303731240030555, | |
| "eval_recall": 0.7259615384615384, | |
| "eval_runtime": 28.5871, | |
| "eval_samples_per_second": 80.036, | |
| "eval_steps_per_second": 10.004, | |
| "step": 19000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 94370, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.8451400467190544e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |