| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.999954477170301, | |
| "global_step": 32949, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.878230691954093e-05, | |
| "loss": 0.7559, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.726019056896709e-05, | |
| "loss": 0.4773, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.573807421839326e-05, | |
| "loss": 0.4477, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.421595786781941e-05, | |
| "loss": 0.4385, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 9.269384151724558e-05, | |
| "loss": 0.4258, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 9.117172516667175e-05, | |
| "loss": 0.419, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 8.964960881609791e-05, | |
| "loss": 0.4017, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 8.812749246552407e-05, | |
| "loss": 0.4165, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 8.660537611495024e-05, | |
| "loss": 0.4148, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 8.508325976437639e-05, | |
| "loss": 0.406, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 8.356114341380256e-05, | |
| "loss": 0.395, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 8.203902706322871e-05, | |
| "loss": 0.3927, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 8.051691071265488e-05, | |
| "loss": 0.3776, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 7.899479436208103e-05, | |
| "loss": 0.3555, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 7.74726780115072e-05, | |
| "loss": 0.3731, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 7.595056166093336e-05, | |
| "loss": 0.3593, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.442844531035952e-05, | |
| "loss": 0.3643, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 7.290632895978569e-05, | |
| "loss": 0.3607, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.138421260921186e-05, | |
| "loss": 0.3404, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 6.986209625863801e-05, | |
| "loss": 0.3527, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 6.833997990806418e-05, | |
| "loss": 0.3513, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7714, | |
| "eval_loss": 0.3165387809276581, | |
| "eval_runtime": 145.7923, | |
| "eval_samples_per_second": 33.52, | |
| "eval_steps_per_second": 2.099, | |
| "step": 10983 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.681786355749033e-05, | |
| "loss": 0.3399, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 6.52957472069165e-05, | |
| "loss": 0.2111, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 6.377363085634266e-05, | |
| "loss": 0.2117, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 6.225151450576882e-05, | |
| "loss": 0.2152, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 6.0729398155194985e-05, | |
| "loss": 0.2217, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 5.920728180462115e-05, | |
| "loss": 0.2194, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 5.768516545404731e-05, | |
| "loss": 0.224, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 5.6163049103473475e-05, | |
| "loss": 0.2262, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 5.464093275289963e-05, | |
| "loss": 0.2308, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 5.31188164023258e-05, | |
| "loss": 0.2266, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 5.159670005175196e-05, | |
| "loss": 0.2162, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 5.0074583701178126e-05, | |
| "loss": 0.2288, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 4.855246735060428e-05, | |
| "loss": 0.2268, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 4.703035100003044e-05, | |
| "loss": 0.2136, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 4.55082346494566e-05, | |
| "loss": 0.2199, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 4.398611829888277e-05, | |
| "loss": 0.2187, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 4.246400194830893e-05, | |
| "loss": 0.2104, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 4.094188559773509e-05, | |
| "loss": 0.2093, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 3.941976924716125e-05, | |
| "loss": 0.2133, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.7897652896587414e-05, | |
| "loss": 0.2122, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 3.6375536546013575e-05, | |
| "loss": 0.2103, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 3.485342019543974e-05, | |
| "loss": 0.2109, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7986, | |
| "eval_loss": 0.33286044001579285, | |
| "eval_runtime": 145.5877, | |
| "eval_samples_per_second": 33.567, | |
| "eval_steps_per_second": 2.102, | |
| "step": 21966 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.3331303844865904e-05, | |
| "loss": 0.138, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 3.1809187494292065e-05, | |
| "loss": 0.104, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 3.0287071143718226e-05, | |
| "loss": 0.112, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 2.876495479314439e-05, | |
| "loss": 0.1109, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.724283844257055e-05, | |
| "loss": 0.1059, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 2.5720722091996712e-05, | |
| "loss": 0.1048, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 2.4198605741422876e-05, | |
| "loss": 0.1067, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 2.2676489390849037e-05, | |
| "loss": 0.1168, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 2.11543730402752e-05, | |
| "loss": 0.1084, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 1.9632256689701363e-05, | |
| "loss": 0.1146, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 1.8110140339127524e-05, | |
| "loss": 0.107, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 1.6588023988553685e-05, | |
| "loss": 0.1107, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 1.5065907637979847e-05, | |
| "loss": 0.1087, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 1.354379128740601e-05, | |
| "loss": 0.1109, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 1.2021674936832173e-05, | |
| "loss": 0.101, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 1.0499558586258334e-05, | |
| "loss": 0.1066, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 8.977442235684496e-06, | |
| "loss": 0.1092, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 7.455325885110659e-06, | |
| "loss": 0.1006, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 5.93320953453682e-06, | |
| "loss": 0.1155, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 4.411093183962983e-06, | |
| "loss": 0.1204, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 2.8889768333891442e-06, | |
| "loss": 0.0972, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 1.3668604828153065e-06, | |
| "loss": 0.0929, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8023, | |
| "eval_loss": 0.438219279050827, | |
| "eval_runtime": 146.5435, | |
| "eval_samples_per_second": 33.348, | |
| "eval_steps_per_second": 2.088, | |
| "step": 32949 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 32949, | |
| "total_flos": 5.70697807036416e+17, | |
| "train_loss": 0.0359802827722491, | |
| "train_runtime": 7164.0026, | |
| "train_samples_per_second": 36.795, | |
| "train_steps_per_second": 4.599 | |
| } | |
| ], | |
| "max_steps": 32949, | |
| "num_train_epochs": 3, | |
| "total_flos": 5.70697807036416e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |