| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.7894608974456787, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5472, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7343358395989975, | |
| "eval_f1": 0.6338815789473684, | |
| "eval_loss": 0.4992983341217041, | |
| "eval_precision": 0.6726405580300865, | |
| "eval_recall": 0.6245226404800873, | |
| "eval_runtime": 5.1542, | |
| "eval_samples_per_second": 77.413, | |
| "eval_steps_per_second": 9.701, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.12828254699707, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4484, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7944862155388471, | |
| "eval_f1": 0.7744208494208494, | |
| "eval_loss": 0.4156816005706787, | |
| "eval_precision": 0.765545388374753, | |
| "eval_recall": 0.8096017457719585, | |
| "eval_runtime": 5.1518, | |
| "eval_samples_per_second": 77.449, | |
| "eval_steps_per_second": 9.705, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.4368088245391846, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3338, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8179269882659713, | |
| "eval_loss": 0.32789668440818787, | |
| "eval_precision": 0.8510239760239761, | |
| "eval_recall": 0.7981905801054737, | |
| "eval_runtime": 5.107, | |
| "eval_samples_per_second": 78.127, | |
| "eval_steps_per_second": 9.79, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.854862689971924, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2902, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8359744037230948, | |
| "eval_loss": 0.30365845561027527, | |
| "eval_precision": 0.8448835433371515, | |
| "eval_recall": 0.828514275322786, | |
| "eval_runtime": 5.1136, | |
| "eval_samples_per_second": 78.028, | |
| "eval_steps_per_second": 9.778, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.4754557609558105, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2756, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8480717680029244, | |
| "eval_loss": 0.292182594537735, | |
| "eval_precision": 0.8498775260257195, | |
| "eval_recall": 0.8463356973995272, | |
| "eval_runtime": 5.1797, | |
| "eval_samples_per_second": 77.031, | |
| "eval_steps_per_second": 9.653, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.3565938472747803, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2514, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8446181767415888, | |
| "eval_loss": 0.3059082329273224, | |
| "eval_precision": 0.835902201887332, | |
| "eval_recall": 0.8560192762320422, | |
| "eval_runtime": 5.076, | |
| "eval_samples_per_second": 78.606, | |
| "eval_steps_per_second": 9.85, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.7644256949424744, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2338, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8347043853938569, | |
| "eval_loss": 0.2969984710216522, | |
| "eval_precision": 0.8277993283927745, | |
| "eval_recall": 0.8431987634115294, | |
| "eval_runtime": 5.1872, | |
| "eval_samples_per_second": 76.92, | |
| "eval_steps_per_second": 9.639, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.6158833503723145, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2205, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8531398028421457, | |
| "eval_loss": 0.29671400785446167, | |
| "eval_precision": 0.8783539291322455, | |
| "eval_recall": 0.835924713584288, | |
| "eval_runtime": 5.0776, | |
| "eval_samples_per_second": 78.581, | |
| "eval_steps_per_second": 9.847, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 7.319582462310791, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2153, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8401647707947546, | |
| "eval_loss": 0.29821664094924927, | |
| "eval_precision": 0.8393298751432535, | |
| "eval_recall": 0.8410165484633569, | |
| "eval_runtime": 5.1535, | |
| "eval_samples_per_second": 77.423, | |
| "eval_steps_per_second": 9.702, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.4924864768981934, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1969, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8377065410088949, | |
| "eval_loss": 0.2942851483821869, | |
| "eval_precision": 0.8423344947735192, | |
| "eval_recall": 0.8335151845790143, | |
| "eval_runtime": 5.1062, | |
| "eval_samples_per_second": 78.14, | |
| "eval_steps_per_second": 9.792, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.3620827198028564, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.185, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8375505157126486, | |
| "eval_loss": 0.29727619886398315, | |
| "eval_precision": 0.8359243697478991, | |
| "eval_recall": 0.8392434988179669, | |
| "eval_runtime": 5.1573, | |
| "eval_samples_per_second": 77.366, | |
| "eval_steps_per_second": 9.695, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 5.206456661224365, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1733, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8377065410088949, | |
| "eval_loss": 0.3074239492416382, | |
| "eval_precision": 0.8423344947735192, | |
| "eval_recall": 0.8335151845790143, | |
| "eval_runtime": 5.1717, | |
| "eval_samples_per_second": 77.151, | |
| "eval_steps_per_second": 9.668, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 2.582509994506836, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1616, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8403508771929824, | |
| "eval_loss": 0.31861984729766846, | |
| "eval_precision": 0.8460491741741742, | |
| "eval_recall": 0.8352882342244045, | |
| "eval_runtime": 5.2202, | |
| "eval_samples_per_second": 76.433, | |
| "eval_steps_per_second": 9.578, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 4.573276519775391, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.16, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8347043853938569, | |
| "eval_loss": 0.32218077778816223, | |
| "eval_precision": 0.8277993283927745, | |
| "eval_recall": 0.8431987634115294, | |
| "eval_runtime": 5.1221, | |
| "eval_samples_per_second": 77.898, | |
| "eval_steps_per_second": 9.762, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 4.897996425628662, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1494, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8464912280701754, | |
| "eval_loss": 0.3260069787502289, | |
| "eval_precision": 0.8522897897897899, | |
| "eval_recall": 0.8413347881432988, | |
| "eval_runtime": 5.1376, | |
| "eval_samples_per_second": 77.663, | |
| "eval_steps_per_second": 9.732, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.7224879264831543, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1501, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8375505157126486, | |
| "eval_loss": 0.32325395941734314, | |
| "eval_precision": 0.8359243697478991, | |
| "eval_recall": 0.8392434988179669, | |
| "eval_runtime": 5.1275, | |
| "eval_samples_per_second": 77.816, | |
| "eval_steps_per_second": 9.751, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 3.4223344326019287, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1468, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8385441718775052, | |
| "eval_loss": 0.32956239581108093, | |
| "eval_precision": 0.8412280701754387, | |
| "eval_recall": 0.8360156392071285, | |
| "eval_runtime": 5.1268, | |
| "eval_samples_per_second": 77.826, | |
| "eval_steps_per_second": 9.753, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.574972152709961, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1423, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8342105263157895, | |
| "eval_loss": 0.3366738557815552, | |
| "eval_precision": 0.8398085585585586, | |
| "eval_recall": 0.82924168030551, | |
| "eval_runtime": 5.1034, | |
| "eval_samples_per_second": 78.183, | |
| "eval_steps_per_second": 9.797, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 7.210375785827637, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1327, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8419946387230413, | |
| "eval_loss": 0.3395027816295624, | |
| "eval_precision": 0.8437691365584814, | |
| "eval_recall": 0.8402891434806329, | |
| "eval_runtime": 5.1186, | |
| "eval_samples_per_second": 77.952, | |
| "eval_steps_per_second": 9.768, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.7272565364837646, | |
| "learning_rate": 0.0, | |
| "loss": 0.1413, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8438308224802573, | |
| "eval_loss": 0.34336015582084656, | |
| "eval_precision": 0.8485409407665505, | |
| "eval_recall": 0.8395617384979087, | |
| "eval_runtime": 5.1504, | |
| "eval_samples_per_second": 77.47, | |
| "eval_steps_per_second": 9.708, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.22777412445818793, | |
| "train_runtime": 1952.8082, | |
| "train_samples_per_second": 37.259, | |
| "train_steps_per_second": 1.249 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 8444128359504000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |