| { | |
| "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": 5.584501266479492, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5636, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7243107769423559, | |
| "eval_f1": 0.6354408930825969, | |
| "eval_loss": 0.507041335105896, | |
| "eval_precision": 0.6575309104533533, | |
| "eval_recall": 0.6274322604109838, | |
| "eval_runtime": 4.6793, | |
| "eval_samples_per_second": 85.269, | |
| "eval_steps_per_second": 10.685, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.185054302215576, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5128, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7343358395989975, | |
| "eval_f1": 0.6976408350014298, | |
| "eval_loss": 0.501506507396698, | |
| "eval_precision": 0.6911057692307692, | |
| "eval_recall": 0.7120385524640844, | |
| "eval_runtime": 5.0462, | |
| "eval_samples_per_second": 79.069, | |
| "eval_steps_per_second": 9.908, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.350452423095703, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4941, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7468671679197995, | |
| "eval_f1": 0.6969020059116857, | |
| "eval_loss": 0.470926433801651, | |
| "eval_precision": 0.6955197132616487, | |
| "eval_recall": 0.698399709038007, | |
| "eval_runtime": 5.0701, | |
| "eval_samples_per_second": 78.696, | |
| "eval_steps_per_second": 9.862, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.6379525661468506, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4702, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7744360902255639, | |
| "eval_f1": 0.7207099303135889, | |
| "eval_loss": 0.4496114253997803, | |
| "eval_precision": 0.7274676737933683, | |
| "eval_recall": 0.7154028005091835, | |
| "eval_runtime": 5.0568, | |
| "eval_samples_per_second": 78.904, | |
| "eval_steps_per_second": 9.888, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.561305046081543, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4704, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7719298245614035, | |
| "eval_f1": 0.7320072332730561, | |
| "eval_loss": 0.45210200548171997, | |
| "eval_precision": 0.7269805119926199, | |
| "eval_recall": 0.7386342971449354, | |
| "eval_runtime": 5.0475, | |
| "eval_samples_per_second": 79.049, | |
| "eval_steps_per_second": 9.906, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.5164988040924072, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.4616, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7644110275689223, | |
| "eval_f1": 0.7212527498662227, | |
| "eval_loss": 0.4490063786506653, | |
| "eval_precision": 0.7175328467153285, | |
| "eval_recall": 0.7258137843244226, | |
| "eval_runtime": 5.059, | |
| "eval_samples_per_second": 78.869, | |
| "eval_steps_per_second": 9.883, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.614987373352051, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.4543, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7819548872180451, | |
| "eval_f1": 0.7449392712550607, | |
| "eval_loss": 0.4381465017795563, | |
| "eval_precision": 0.7389190734915643, | |
| "eval_recall": 0.7532278596108384, | |
| "eval_runtime": 5.0493, | |
| "eval_samples_per_second": 79.021, | |
| "eval_steps_per_second": 9.902, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.577010631561279, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4532, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8070175438596491, | |
| "eval_f1": 0.7519478107817887, | |
| "eval_loss": 0.41970905661582947, | |
| "eval_precision": 0.7744157656086501, | |
| "eval_recall": 0.7384524458992544, | |
| "eval_runtime": 5.0679, | |
| "eval_samples_per_second": 78.731, | |
| "eval_steps_per_second": 9.866, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 3.4122800827026367, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.4517, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7969924812030075, | |
| "eval_f1": 0.7544988415815261, | |
| "eval_loss": 0.4194534420967102, | |
| "eval_precision": 0.7551480443523821, | |
| "eval_recall": 0.753864338970722, | |
| "eval_runtime": 5.0418, | |
| "eval_samples_per_second": 79.138, | |
| "eval_steps_per_second": 9.917, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.07973575592041, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.4438, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8170426065162907, | |
| "eval_f1": 0.7539886990379824, | |
| "eval_loss": 0.41020727157592773, | |
| "eval_precision": 0.8012931034482759, | |
| "eval_recall": 0.7330423713402436, | |
| "eval_runtime": 5.0907, | |
| "eval_samples_per_second": 78.378, | |
| "eval_steps_per_second": 9.822, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 3.512185573577881, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.4389, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_f1": 0.7875973059089472, | |
| "eval_loss": 0.41120436787605286, | |
| "eval_precision": 0.7933469644542308, | |
| "eval_recall": 0.7826422985997454, | |
| "eval_runtime": 5.0564, | |
| "eval_samples_per_second": 78.909, | |
| "eval_steps_per_second": 9.888, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 6.24143123626709, | |
| "learning_rate": 2e-05, | |
| "loss": 0.4428, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7969924812030075, | |
| "eval_f1": 0.7603603603603604, | |
| "eval_loss": 0.417868971824646, | |
| "eval_precision": 0.7554945054945055, | |
| "eval_recall": 0.766366612111293, | |
| "eval_runtime": 5.0497, | |
| "eval_samples_per_second": 79.014, | |
| "eval_steps_per_second": 9.901, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.4772980213165283, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.4421, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.7828368575397437, | |
| "eval_loss": 0.4030059576034546, | |
| "eval_precision": 0.810950998442147, | |
| "eval_recall": 0.7661847608656118, | |
| "eval_runtime": 5.0599, | |
| "eval_samples_per_second": 78.855, | |
| "eval_steps_per_second": 9.882, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 4.054733753204346, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.4403, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.7915149151881459, | |
| "eval_loss": 0.4036868214607239, | |
| "eval_precision": 0.801371627277996, | |
| "eval_recall": 0.7836879432624113, | |
| "eval_runtime": 5.0499, | |
| "eval_samples_per_second": 79.012, | |
| "eval_steps_per_second": 9.901, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 2.8372957706451416, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.4392, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8220551378446115, | |
| "eval_f1": 0.7858811080458032, | |
| "eval_loss": 0.4077085256576538, | |
| "eval_precision": 0.7851800470474697, | |
| "eval_recall": 0.7865975631933079, | |
| "eval_runtime": 5.0546, | |
| "eval_samples_per_second": 78.938, | |
| "eval_steps_per_second": 9.892, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 3.2562994956970215, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4329, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8195488721804511, | |
| "eval_f1": 0.7834006876168647, | |
| "eval_loss": 0.40617087483406067, | |
| "eval_precision": 0.7820378151260504, | |
| "eval_recall": 0.7848245135479177, | |
| "eval_runtime": 5.0513, | |
| "eval_samples_per_second": 78.99, | |
| "eval_steps_per_second": 9.898, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 3.811532497406006, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.4338, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8145363408521303, | |
| "eval_f1": 0.7773840400506664, | |
| "eval_loss": 0.4058062434196472, | |
| "eval_precision": 0.7760504201680672, | |
| "eval_recall": 0.7787779596290234, | |
| "eval_runtime": 5.0681, | |
| "eval_samples_per_second": 78.728, | |
| "eval_steps_per_second": 9.866, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.1403231620788574, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4407, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8145363408521303, | |
| "eval_f1": 0.7762775050009092, | |
| "eval_loss": 0.40417465567588806, | |
| "eval_precision": 0.7762775050009092, | |
| "eval_recall": 0.7762775050009092, | |
| "eval_runtime": 5.0621, | |
| "eval_samples_per_second": 78.821, | |
| "eval_steps_per_second": 9.877, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 2.9727184772491455, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.4329, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8195488721804511, | |
| "eval_f1": 0.781223345924211, | |
| "eval_loss": 0.4033084809780121, | |
| "eval_precision": 0.7826852418860992, | |
| "eval_recall": 0.7798236042916894, | |
| "eval_runtime": 5.0559, | |
| "eval_samples_per_second": 78.918, | |
| "eval_steps_per_second": 9.889, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 5.092881202697754, | |
| "learning_rate": 0.0, | |
| "loss": 0.4292, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8145363408521303, | |
| "eval_f1": 0.7762775050009092, | |
| "eval_loss": 0.40420523285865784, | |
| "eval_precision": 0.7762775050009092, | |
| "eval_recall": 0.7762775050009092, | |
| "eval_runtime": 5.0503, | |
| "eval_samples_per_second": 79.006, | |
| "eval_steps_per_second": 9.9, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7588990440528000.0, | |
| "train_loss": 0.45743675075593543, | |
| "train_runtime": 1951.2471, | |
| "train_samples_per_second": 37.289, | |
| "train_steps_per_second": 1.25 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7588990440528000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |