| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9465930018416207, | |
| "global_step": 24000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.9590750971966443e-05, | |
| "loss": 2.4744, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.9181501943932885e-05, | |
| "loss": 1.9135, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.8772252915899326e-05, | |
| "loss": 1.796, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.8363003887865768e-05, | |
| "loss": 1.7038, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.795375485983221e-05, | |
| "loss": 1.6516, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 1.754450583179865e-05, | |
| "loss": 1.6246, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1.7135256803765093e-05, | |
| "loss": 1.5577, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.6726007775731534e-05, | |
| "loss": 1.5208, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.6316758747697976e-05, | |
| "loss": 1.5118, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.5907509719664418e-05, | |
| "loss": 1.5089, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.5498260691630856e-05, | |
| "loss": 1.4391, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1.50890116635973e-05, | |
| "loss": 1.4344, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 1.467976263556374e-05, | |
| "loss": 1.4238, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.4270513607530182e-05, | |
| "loss": 1.4384, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.3861264579496626e-05, | |
| "loss": 1.3665, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.3452015551463067e-05, | |
| "loss": 1.3999, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_em": 0.5297689323663349, | |
| "eval_f1": 0.6106438572272215, | |
| "eval_loss": 1.3775073289871216, | |
| "eval_runtime": 131.0166, | |
| "eval_samples_per_second": 90.508, | |
| "eval_steps_per_second": 5.663, | |
| "step": 8145 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 1.3042766523429509e-05, | |
| "loss": 1.1856, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.263351749539595e-05, | |
| "loss": 1.0984, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.222426846736239e-05, | |
| "loss": 1.1179, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.1815019439328832e-05, | |
| "loss": 1.0886, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.1405770411295274e-05, | |
| "loss": 1.1104, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 1.0996521383261715e-05, | |
| "loss": 1.1339, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.0587272355228157e-05, | |
| "loss": 1.1361, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.01780233271946e-05, | |
| "loss": 1.1057, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 9.76877429916104e-06, | |
| "loss": 1.0967, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 9.359525271127482e-06, | |
| "loss": 1.1119, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 8.950276243093923e-06, | |
| "loss": 1.0718, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 8.541027215060365e-06, | |
| "loss": 1.093, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 8.131778187026806e-06, | |
| "loss": 1.0782, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 7.722529158993248e-06, | |
| "loss": 1.0655, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 7.313280130959689e-06, | |
| "loss": 1.0953, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 6.904031102926131e-06, | |
| "loss": 1.0765, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_em": 0.5461291954798448, | |
| "eval_f1": 0.6263308103966033, | |
| "eval_loss": 1.385632038116455, | |
| "eval_runtime": 134.345, | |
| "eval_samples_per_second": 88.265, | |
| "eval_steps_per_second": 5.523, | |
| "step": 16290 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 6.494782074892573e-06, | |
| "loss": 0.9839, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 6.0855330468590145e-06, | |
| "loss": 0.8102, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 5.676284018825455e-06, | |
| "loss": 0.8344, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 5.267034990791897e-06, | |
| "loss": 0.8254, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 4.857785962758339e-06, | |
| "loss": 0.8348, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 4.44853693472478e-06, | |
| "loss": 0.8377, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 4.0392879066912225e-06, | |
| "loss": 0.8376, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 3.6300388786576637e-06, | |
| "loss": 0.8231, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 3.220789850624105e-06, | |
| "loss": 0.8427, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 2.8115408225905465e-06, | |
| "loss": 0.8219, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 2.402291794556988e-06, | |
| "loss": 0.8272, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 1.9930427665234297e-06, | |
| "loss": 0.8353, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 1.5837937384898713e-06, | |
| "loss": 0.8305, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 1.1745447104563129e-06, | |
| "loss": 0.8231, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 7.652956824227544e-07, | |
| "loss": 0.8398, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 3.5604665438919586e-07, | |
| "loss": 0.8243, | |
| "step": 24000 | |
| } | |
| ], | |
| "max_steps": 24435, | |
| "num_train_epochs": 3, | |
| "total_flos": 7.525072232402227e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |