| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 10560, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.1869616508483887, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.7039, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9123024942859982, | |
| "eval_f1": 0.5240871432954894, | |
| "eval_loss": 0.3293141722679138, | |
| "eval_precision": 0.5552665799739922, | |
| "eval_recall": 0.4962231260894829, | |
| "eval_runtime": 4.4474, | |
| "eval_samples_per_second": 210.236, | |
| "eval_steps_per_second": 3.373, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.910118818283081, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.2536, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9424127993640068, | |
| "eval_f1": 0.7274131274131275, | |
| "eval_loss": 0.1835452765226364, | |
| "eval_precision": 0.6529574861367837, | |
| "eval_recall": 0.821034282393957, | |
| "eval_runtime": 4.5333, | |
| "eval_samples_per_second": 206.253, | |
| "eval_steps_per_second": 3.309, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 10.446182250976562, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.1831, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9440027824704362, | |
| "eval_f1": 0.7365129007036748, | |
| "eval_loss": 0.18317362666130066, | |
| "eval_precision": 0.6677693761814745, | |
| "eval_recall": 0.821034282393957, | |
| "eval_runtime": 4.5013, | |
| "eval_samples_per_second": 207.715, | |
| "eval_steps_per_second": 3.332, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 5.82688045501709, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1623, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9535426811090132, | |
| "eval_f1": 0.7789361133386795, | |
| "eval_loss": 0.14634285867214203, | |
| "eval_precision": 0.7212871287128713, | |
| "eval_recall": 0.8466008134805346, | |
| "eval_runtime": 4.5285, | |
| "eval_samples_per_second": 206.47, | |
| "eval_steps_per_second": 3.312, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.118807077407837, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.1439, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9541389247739243, | |
| "eval_f1": 0.7746591820368887, | |
| "eval_loss": 0.1386542022228241, | |
| "eval_precision": 0.7173267326732673, | |
| "eval_recall": 0.841952353282975, | |
| "eval_runtime": 4.5083, | |
| "eval_samples_per_second": 207.394, | |
| "eval_steps_per_second": 3.327, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.6887811422348022, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.1348, | |
| "step": 3168 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9553314121037464, | |
| "eval_f1": 0.7892923403127484, | |
| "eval_loss": 0.1382703185081482, | |
| "eval_precision": 0.7256335282651072, | |
| "eval_recall": 0.8651946542707728, | |
| "eval_runtime": 4.5098, | |
| "eval_samples_per_second": 207.326, | |
| "eval_steps_per_second": 3.326, | |
| "step": 3168 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 8.010842323303223, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.1293, | |
| "step": 3696 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9545364205505317, | |
| "eval_f1": 0.7884564469155414, | |
| "eval_loss": 0.1393628716468811, | |
| "eval_precision": 0.7242217898832685, | |
| "eval_recall": 0.8651946542707728, | |
| "eval_runtime": 4.5377, | |
| "eval_samples_per_second": 206.052, | |
| "eval_steps_per_second": 3.306, | |
| "step": 3696 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.8240045309066772, | |
| "learning_rate": 3e-05, | |
| "loss": 0.124, | |
| "step": 4224 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9569213952101759, | |
| "eval_f1": 0.7969124301304232, | |
| "eval_loss": 0.13507381081581116, | |
| "eval_precision": 0.7352652259332023, | |
| "eval_recall": 0.8698431144683324, | |
| "eval_runtime": 4.5382, | |
| "eval_samples_per_second": 206.027, | |
| "eval_steps_per_second": 3.305, | |
| "step": 4224 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.0099204778671265, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.1176, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9561264036569611, | |
| "eval_f1": 0.7929824561403509, | |
| "eval_loss": 0.13042420148849487, | |
| "eval_precision": 0.7404233870967742, | |
| "eval_recall": 0.8535735037768739, | |
| "eval_runtime": 4.5394, | |
| "eval_samples_per_second": 205.976, | |
| "eval_steps_per_second": 3.304, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.6231740713119507, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1153, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9585113783166054, | |
| "eval_f1": 0.8051240119923686, | |
| "eval_loss": 0.1277889907360077, | |
| "eval_precision": 0.7582135523613963, | |
| "eval_recall": 0.8582219639744335, | |
| "eval_runtime": 4.5204, | |
| "eval_samples_per_second": 206.838, | |
| "eval_steps_per_second": 3.318, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.4694974422454834, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.111, | |
| "step": 5808 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9579151346516943, | |
| "eval_f1": 0.7968959058067969, | |
| "eval_loss": 0.13041619956493378, | |
| "eval_precision": 0.7385912698412699, | |
| "eval_recall": 0.8651946542707728, | |
| "eval_runtime": 4.5135, | |
| "eval_samples_per_second": 207.154, | |
| "eval_steps_per_second": 3.323, | |
| "step": 5808 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.5000303983688354, | |
| "learning_rate": 2e-05, | |
| "loss": 0.109, | |
| "step": 6336 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9565238994335685, | |
| "eval_f1": 0.7986055242692411, | |
| "eval_loss": 0.13234680891036987, | |
| "eval_precision": 0.7415338645418327, | |
| "eval_recall": 0.8651946542707728, | |
| "eval_runtime": 4.5427, | |
| "eval_samples_per_second": 205.826, | |
| "eval_steps_per_second": 3.302, | |
| "step": 6336 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.91750168800354, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1077, | |
| "step": 6864 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9597038656464275, | |
| "eval_f1": 0.812959433705418, | |
| "eval_loss": 0.12529084086418152, | |
| "eval_precision": 0.7648565573770492, | |
| "eval_recall": 0.8675188843695526, | |
| "eval_runtime": 4.5154, | |
| "eval_samples_per_second": 207.07, | |
| "eval_steps_per_second": 3.322, | |
| "step": 6864 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 1.4095473289489746, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1032, | |
| "step": 7392 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9593063698698201, | |
| "eval_f1": 0.8103683492496588, | |
| "eval_loss": 0.1242511197924614, | |
| "eval_precision": 0.7638888888888888, | |
| "eval_recall": 0.862870424171993, | |
| "eval_runtime": 4.5063, | |
| "eval_samples_per_second": 207.489, | |
| "eval_steps_per_second": 3.329, | |
| "step": 7392 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.268137812614441, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1035, | |
| "step": 7920 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9597038656464275, | |
| "eval_f1": 0.8138457345325703, | |
| "eval_loss": 0.1261007934808731, | |
| "eval_precision": 0.7664271047227926, | |
| "eval_recall": 0.8675188843695526, | |
| "eval_runtime": 4.55, | |
| "eval_samples_per_second": 205.495, | |
| "eval_steps_per_second": 3.297, | |
| "step": 7920 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.6868817806243896, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1017, | |
| "step": 8448 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9577163867633907, | |
| "eval_f1": 0.7977254264825344, | |
| "eval_loss": 0.12578710913658142, | |
| "eval_precision": 0.7469574036511156, | |
| "eval_recall": 0.8558977338756537, | |
| "eval_runtime": 4.5309, | |
| "eval_samples_per_second": 206.363, | |
| "eval_steps_per_second": 3.311, | |
| "step": 8448 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 1.2091130018234253, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1004, | |
| "step": 8976 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9589088740932128, | |
| "eval_f1": 0.809845820935894, | |
| "eval_loss": 0.12778569757938385, | |
| "eval_precision": 0.7575910931174089, | |
| "eval_recall": 0.8698431144683324, | |
| "eval_runtime": 4.5586, | |
| "eval_samples_per_second": 205.105, | |
| "eval_steps_per_second": 3.29, | |
| "step": 8976 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 1.5732147693634033, | |
| "learning_rate": 5e-06, | |
| "loss": 0.099, | |
| "step": 9504 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9585113783166054, | |
| "eval_f1": 0.8050687516850902, | |
| "eval_loss": 0.12837953865528107, | |
| "eval_precision": 0.7510060362173038, | |
| "eval_recall": 0.8675188843695526, | |
| "eval_runtime": 4.513, | |
| "eval_samples_per_second": 207.18, | |
| "eval_steps_per_second": 3.324, | |
| "step": 9504 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.611748218536377, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0991, | |
| "step": 10032 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.958113882539998, | |
| "eval_f1": 0.8055480010878433, | |
| "eval_loss": 0.12564244866371155, | |
| "eval_precision": 0.7571574642126789, | |
| "eval_recall": 0.8605461940732132, | |
| "eval_runtime": 4.5504, | |
| "eval_samples_per_second": 205.478, | |
| "eval_steps_per_second": 3.296, | |
| "step": 10032 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.8662045001983643, | |
| "learning_rate": 0.0, | |
| "loss": 0.0984, | |
| "step": 10560 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9591076219815164, | |
| "eval_f1": 0.812483039348711, | |
| "eval_loss": 0.12664274871349335, | |
| "eval_precision": 0.7622199592668024, | |
| "eval_recall": 0.8698431144683324, | |
| "eval_runtime": 4.5353, | |
| "eval_samples_per_second": 206.161, | |
| "eval_steps_per_second": 3.307, | |
| "step": 10560 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 10560, | |
| "total_flos": 4552961808488766.0, | |
| "train_loss": 0.1550482457334345, | |
| "train_runtime": 1230.1483, | |
| "train_samples_per_second": 137.17, | |
| "train_steps_per_second": 8.584 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10560, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 4552961808488766.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |