| { | |
| "best_metric": 1.1290708780288696, | |
| "best_model_checkpoint": "/home/ubuntu/ML/source_code/fine-tuned-legalroberta/checkpoint-17000", | |
| "epoch": 4.779308405960078, | |
| "global_step": 17000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.9437728422828227e-05, | |
| "loss": 1.5399, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 1.5552507638931274, | |
| "eval_runtime": 2.1088, | |
| "eval_samples_per_second": 72.078, | |
| "eval_steps_per_second": 7.587, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.8875456845656453e-05, | |
| "loss": 1.3513, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 1.3741689920425415, | |
| "eval_runtime": 2.1146, | |
| "eval_samples_per_second": 71.881, | |
| "eval_steps_per_second": 7.566, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 1.831318526848468e-05, | |
| "loss": 1.2809, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 1.3449822664260864, | |
| "eval_runtime": 2.1138, | |
| "eval_samples_per_second": 71.909, | |
| "eval_steps_per_second": 7.569, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 1.7750913691312905e-05, | |
| "loss": 1.2484, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_loss": 1.3560587167739868, | |
| "eval_runtime": 2.1078, | |
| "eval_samples_per_second": 72.113, | |
| "eval_steps_per_second": 7.591, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.718864211414113e-05, | |
| "loss": 1.2016, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 1.3128280639648438, | |
| "eval_runtime": 2.1023, | |
| "eval_samples_per_second": 72.303, | |
| "eval_steps_per_second": 7.611, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 1.662637053696936e-05, | |
| "loss": 1.1724, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 1.3004732131958008, | |
| "eval_runtime": 2.1125, | |
| "eval_samples_per_second": 71.951, | |
| "eval_steps_per_second": 7.574, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.606409895979758e-05, | |
| "loss": 1.1664, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_loss": 1.2773096561431885, | |
| "eval_runtime": 2.119, | |
| "eval_samples_per_second": 71.732, | |
| "eval_steps_per_second": 7.551, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 1.550182738262581e-05, | |
| "loss": 1.1399, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_loss": 1.2295717000961304, | |
| "eval_runtime": 2.119, | |
| "eval_samples_per_second": 71.733, | |
| "eval_steps_per_second": 7.551, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1.4939555805454037e-05, | |
| "loss": 1.1136, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 1.2814128398895264, | |
| "eval_runtime": 2.1037, | |
| "eval_samples_per_second": 72.255, | |
| "eval_steps_per_second": 7.606, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.4377284228282262e-05, | |
| "loss": 1.1026, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_loss": 1.2653945684432983, | |
| "eval_runtime": 2.1041, | |
| "eval_samples_per_second": 72.24, | |
| "eval_steps_per_second": 7.604, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.3815012651110486e-05, | |
| "loss": 1.0943, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_loss": 1.210158109664917, | |
| "eval_runtime": 2.1061, | |
| "eval_samples_per_second": 72.17, | |
| "eval_steps_per_second": 7.597, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 1.3252741073938714e-05, | |
| "loss": 1.0768, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "eval_loss": 1.1943004131317139, | |
| "eval_runtime": 2.1055, | |
| "eval_samples_per_second": 72.191, | |
| "eval_steps_per_second": 7.599, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.269046949676694e-05, | |
| "loss": 1.0741, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_loss": 1.2276432514190674, | |
| "eval_runtime": 2.1096, | |
| "eval_samples_per_second": 72.05, | |
| "eval_steps_per_second": 7.584, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 1.2128197919595167e-05, | |
| "loss": 1.0561, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 1.1935926675796509, | |
| "eval_runtime": 2.1073, | |
| "eval_samples_per_second": 72.129, | |
| "eval_steps_per_second": 7.593, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 1.1565926342423393e-05, | |
| "loss": 1.0384, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 1.2100528478622437, | |
| "eval_runtime": 2.1142, | |
| "eval_samples_per_second": 71.894, | |
| "eval_steps_per_second": 7.568, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.1003654765251617e-05, | |
| "loss": 1.037, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_loss": 1.1716400384902954, | |
| "eval_runtime": 2.1168, | |
| "eval_samples_per_second": 71.805, | |
| "eval_steps_per_second": 7.558, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 1.0441383188079844e-05, | |
| "loss": 1.0295, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_loss": 1.1552788019180298, | |
| "eval_runtime": 2.114, | |
| "eval_samples_per_second": 71.901, | |
| "eval_steps_per_second": 7.569, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 9.87911161090807e-06, | |
| "loss": 1.0255, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 1.211492896080017, | |
| "eval_runtime": 2.1048, | |
| "eval_samples_per_second": 72.217, | |
| "eval_steps_per_second": 7.602, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 9.316840033736295e-06, | |
| "loss": 1.0207, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 1.1855554580688477, | |
| "eval_runtime": 2.1112, | |
| "eval_samples_per_second": 71.998, | |
| "eval_steps_per_second": 7.579, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 8.754568456564521e-06, | |
| "loss": 1.0097, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "eval_loss": 1.1690603494644165, | |
| "eval_runtime": 2.1041, | |
| "eval_samples_per_second": 72.238, | |
| "eval_steps_per_second": 7.604, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 8.192296879392747e-06, | |
| "loss": 0.9996, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "eval_loss": 1.1681824922561646, | |
| "eval_runtime": 2.107, | |
| "eval_samples_per_second": 72.141, | |
| "eval_steps_per_second": 7.594, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 7.630025302220974e-06, | |
| "loss": 0.9912, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "eval_loss": 1.200119972229004, | |
| "eval_runtime": 2.1358, | |
| "eval_samples_per_second": 71.168, | |
| "eval_steps_per_second": 7.491, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 7.067753725049199e-06, | |
| "loss": 0.9844, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "eval_loss": 1.129595160484314, | |
| "eval_runtime": 2.111, | |
| "eval_samples_per_second": 72.004, | |
| "eval_steps_per_second": 7.579, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 6.505482147877426e-06, | |
| "loss": 0.9908, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "eval_loss": 1.1818418502807617, | |
| "eval_runtime": 2.1165, | |
| "eval_samples_per_second": 71.817, | |
| "eval_steps_per_second": 7.56, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 5.943210570705651e-06, | |
| "loss": 0.9798, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "eval_loss": 1.1727079153060913, | |
| "eval_runtime": 2.1056, | |
| "eval_samples_per_second": 72.188, | |
| "eval_steps_per_second": 7.599, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 5.380938993533877e-06, | |
| "loss": 0.9699, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "eval_loss": 1.2000970840454102, | |
| "eval_runtime": 2.1003, | |
| "eval_samples_per_second": 72.372, | |
| "eval_steps_per_second": 7.618, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 4.818667416362104e-06, | |
| "loss": 0.9706, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "eval_loss": 1.1695332527160645, | |
| "eval_runtime": 2.1004, | |
| "eval_samples_per_second": 72.366, | |
| "eval_steps_per_second": 7.617, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 4.2563958391903294e-06, | |
| "loss": 0.974, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "eval_loss": 1.1403844356536865, | |
| "eval_runtime": 2.1018, | |
| "eval_samples_per_second": 72.318, | |
| "eval_steps_per_second": 7.612, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 3.694124262018555e-06, | |
| "loss": 0.9561, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "eval_loss": 1.1319295167922974, | |
| "eval_runtime": 2.1093, | |
| "eval_samples_per_second": 72.062, | |
| "eval_steps_per_second": 7.586, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 3.1318526848467813e-06, | |
| "loss": 0.9598, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "eval_loss": 1.1552999019622803, | |
| "eval_runtime": 2.116, | |
| "eval_samples_per_second": 71.835, | |
| "eval_steps_per_second": 7.562, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 2.569581107675007e-06, | |
| "loss": 0.9626, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "eval_loss": 1.1351696252822876, | |
| "eval_runtime": 2.1126, | |
| "eval_samples_per_second": 71.95, | |
| "eval_steps_per_second": 7.574, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 2.007309530503233e-06, | |
| "loss": 0.9631, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "eval_loss": 1.181087851524353, | |
| "eval_runtime": 2.1067, | |
| "eval_samples_per_second": 72.151, | |
| "eval_steps_per_second": 7.595, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 1.4450379533314593e-06, | |
| "loss": 0.9485, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "eval_loss": 1.1709717512130737, | |
| "eval_runtime": 2.1077, | |
| "eval_samples_per_second": 72.118, | |
| "eval_steps_per_second": 7.591, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 8.827663761596852e-07, | |
| "loss": 0.946, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "eval_loss": 1.1290708780288696, | |
| "eval_runtime": 2.1091, | |
| "eval_samples_per_second": 72.067, | |
| "eval_steps_per_second": 7.586, | |
| "step": 17000 | |
| } | |
| ], | |
| "max_steps": 17785, | |
| "num_train_epochs": 5, | |
| "total_flos": 4.47488103892009e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |