| { | |
| "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": 4.9823102951049805, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5456, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7468671679197995, | |
| "eval_f1": 0.6641666041651041, | |
| "eval_loss": 0.49161723256111145, | |
| "eval_precision": 0.6912698412698413, | |
| "eval_recall": 0.6533915257319513, | |
| "eval_runtime": 1.7896, | |
| "eval_samples_per_second": 222.952, | |
| "eval_steps_per_second": 27.939, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.931857109069824, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4369, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8120300751879699, | |
| "eval_f1": 0.7912087912087912, | |
| "eval_loss": 0.41080209612846375, | |
| "eval_precision": 0.7802706552706553, | |
| "eval_recall": 0.822013093289689, | |
| "eval_runtime": 1.8221, | |
| "eval_samples_per_second": 218.982, | |
| "eval_steps_per_second": 27.441, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.00280499458313, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3316, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.81524926686217, | |
| "eval_loss": 0.3294452726840973, | |
| "eval_precision": 0.8462682339611953, | |
| "eval_recall": 0.7964175304600837, | |
| "eval_runtime": 1.7904, | |
| "eval_samples_per_second": 222.854, | |
| "eval_steps_per_second": 27.927, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.0830881595611572, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2909, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8499551039516197, | |
| "eval_loss": 0.30187976360321045, | |
| "eval_precision": 0.8547473867595818, | |
| "eval_recall": 0.8456082924168031, | |
| "eval_runtime": 1.8, | |
| "eval_samples_per_second": 221.672, | |
| "eval_steps_per_second": 27.778, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.5169291496276855, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2584, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.842789598108747, | |
| "eval_loss": 0.3023463487625122, | |
| "eval_precision": 0.842789598108747, | |
| "eval_recall": 0.842789598108747, | |
| "eval_runtime": 1.7982, | |
| "eval_samples_per_second": 221.888, | |
| "eval_steps_per_second": 27.805, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.894607424736023, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.237, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8375505157126486, | |
| "eval_loss": 0.30203157663345337, | |
| "eval_precision": 0.8359243697478991, | |
| "eval_recall": 0.8392434988179669, | |
| "eval_runtime": 1.7962, | |
| "eval_samples_per_second": 222.138, | |
| "eval_steps_per_second": 27.837, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.33639559149742126, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2186, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8491157824491158, | |
| "eval_loss": 0.2988818287849426, | |
| "eval_precision": 0.8424908424908425, | |
| "eval_recall": 0.8570649208947081, | |
| "eval_runtime": 1.8005, | |
| "eval_samples_per_second": 221.607, | |
| "eval_steps_per_second": 27.77, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.405568242073059, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2108, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8614765038536611, | |
| "eval_loss": 0.29612118005752563, | |
| "eval_precision": 0.8686536646744258, | |
| "eval_recall": 0.8552009456264775, | |
| "eval_runtime": 1.8282, | |
| "eval_samples_per_second": 218.244, | |
| "eval_steps_per_second": 27.349, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.2904157638549805, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.1898, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8480717680029244, | |
| "eval_loss": 0.30133265256881714, | |
| "eval_precision": 0.8498775260257195, | |
| "eval_recall": 0.8463356973995272, | |
| "eval_runtime": 1.7951, | |
| "eval_samples_per_second": 222.272, | |
| "eval_steps_per_second": 27.854, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.820188522338867, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1894, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8456742372671576, | |
| "eval_loss": 0.32312536239624023, | |
| "eval_precision": 0.8536697247706422, | |
| "eval_recall": 0.8388343335151845, | |
| "eval_runtime": 1.8085, | |
| "eval_samples_per_second": 220.63, | |
| "eval_steps_per_second": 27.648, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.6913771629333496, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1817, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8514869535493182, | |
| "eval_loss": 0.301199346780777, | |
| "eval_precision": 0.8523821128305106, | |
| "eval_recall": 0.8506092016730314, | |
| "eval_runtime": 1.8217, | |
| "eval_samples_per_second": 219.025, | |
| "eval_steps_per_second": 27.447, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 5.912967681884766, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1723, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8359175094431583, | |
| "eval_loss": 0.29794374108314514, | |
| "eval_precision": 0.8376607470912432, | |
| "eval_recall": 0.8342425895617385, | |
| "eval_runtime": 1.7955, | |
| "eval_samples_per_second": 222.218, | |
| "eval_steps_per_second": 27.847, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.4766283929347992, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1547, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8411818738518064, | |
| "eval_loss": 0.2937452793121338, | |
| "eval_precision": 0.8448542607834644, | |
| "eval_recall": 0.8377886888525186, | |
| "eval_runtime": 1.8242, | |
| "eval_samples_per_second": 218.73, | |
| "eval_steps_per_second": 27.41, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 4.687190532684326, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1569, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8485547445255475, | |
| "eval_loss": 0.3064776360988617, | |
| "eval_precision": 0.8383458646616542, | |
| "eval_recall": 0.8627932351336607, | |
| "eval_runtime": 1.7921, | |
| "eval_samples_per_second": 222.641, | |
| "eval_steps_per_second": 27.9, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 4.828568458557129, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1442, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8529524583135901, | |
| "eval_loss": 0.2884095013141632, | |
| "eval_precision": 0.8504480286738352, | |
| "eval_recall": 0.8556101109292599, | |
| "eval_runtime": 1.8008, | |
| "eval_samples_per_second": 221.563, | |
| "eval_steps_per_second": 27.765, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 3.8095035552978516, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1435, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8563025210084034, | |
| "eval_loss": 0.30156683921813965, | |
| "eval_precision": 0.8529936381473334, | |
| "eval_recall": 0.8598836152027641, | |
| "eval_runtime": 1.8479, | |
| "eval_samples_per_second": 215.919, | |
| "eval_steps_per_second": 27.058, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.34168025851249695, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1378, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8556962025316456, | |
| "eval_loss": 0.31144392490386963, | |
| "eval_precision": 0.8477571494464944, | |
| "eval_recall": 0.8656119294417166, | |
| "eval_runtime": 1.8294, | |
| "eval_samples_per_second": 218.106, | |
| "eval_steps_per_second": 27.332, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.2586296498775482, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1377, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8550328105883662, | |
| "eval_loss": 0.3096340000629425, | |
| "eval_precision": 0.8482905982905984, | |
| "eval_recall": 0.8631114748136025, | |
| "eval_runtime": 1.8332, | |
| "eval_samples_per_second": 217.658, | |
| "eval_steps_per_second": 27.275, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.0663389191031456, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1307, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8556004584112431, | |
| "eval_loss": 0.30646833777427673, | |
| "eval_precision": 0.8538865546218487, | |
| "eval_recall": 0.85738316057465, | |
| "eval_runtime": 1.8307, | |
| "eval_samples_per_second": 217.954, | |
| "eval_steps_per_second": 27.313, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.8260014057159424, | |
| "learning_rate": 0.0, | |
| "loss": 0.126, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8656072933585827, | |
| "eval_loss": 0.30540019273757935, | |
| "eval_precision": 0.8614399005740664, | |
| "eval_recall": 0.8702036733951628, | |
| "eval_runtime": 1.8294, | |
| "eval_samples_per_second": 218.099, | |
| "eval_steps_per_second": 27.331, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.21972156196344095, | |
| "train_runtime": 635.9425, | |
| "train_samples_per_second": 114.413, | |
| "train_steps_per_second": 3.837 | |
| } | |
| ], | |
| "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 | |
| } | |