| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 50, | |
| "global_step": 320, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03125, | |
| "grad_norm": 159.0, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 1.1247, | |
| "num_input_tokens_seen": 1465248, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 65.0, | |
| "learning_rate": 2.9987186375809513e-05, | |
| "loss": 0.638, | |
| "num_input_tokens_seen": 2893200, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09375, | |
| "grad_norm": 10.0625, | |
| "learning_rate": 2.984328439990804e-05, | |
| "loss": 0.2887, | |
| "num_input_tokens_seen": 4373616, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 14.875, | |
| "learning_rate": 2.9541003989089956e-05, | |
| "loss": 0.2392, | |
| "num_input_tokens_seen": 5755248, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15625, | |
| "grad_norm": 9.8125, | |
| "learning_rate": 2.9083570487361445e-05, | |
| "loss": 0.226, | |
| "num_input_tokens_seen": 7213808, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 19.0, | |
| "learning_rate": 2.8475864728379682e-05, | |
| "loss": 0.2489, | |
| "num_input_tokens_seen": 8649584, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21875, | |
| "grad_norm": 27.75, | |
| "learning_rate": 2.7724370956763605e-05, | |
| "loss": 0.2462, | |
| "num_input_tokens_seen": 10054416, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 12.5, | |
| "learning_rate": 2.6837107640945904e-05, | |
| "loss": 0.2133, | |
| "num_input_tokens_seen": 11521216, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.28125, | |
| "grad_norm": 8.625, | |
| "learning_rate": 2.5823541915795932e-05, | |
| "loss": 0.2035, | |
| "num_input_tokens_seen": 12912576, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 4.25, | |
| "learning_rate": 2.469448856791411e-05, | |
| "loss": 0.2226, | |
| "num_input_tokens_seen": 14369728, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.34375, | |
| "grad_norm": 22.625, | |
| "learning_rate": 2.3461994641428768e-05, | |
| "loss": 0.2314, | |
| "num_input_tokens_seen": 15796256, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 19.875, | |
| "learning_rate": 2.2139210895556104e-05, | |
| "loss": 0.2564, | |
| "num_input_tokens_seen": 17240400, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.40625, | |
| "grad_norm": 7.78125, | |
| "learning_rate": 2.074025148547635e-05, | |
| "loss": 0.2442, | |
| "num_input_tokens_seen": 18663216, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4375, | |
| "grad_norm": 12.125, | |
| "learning_rate": 1.928004336373658e-05, | |
| "loss": 0.2115, | |
| "num_input_tokens_seen": 20069056, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.46875, | |
| "grad_norm": 21.0, | |
| "learning_rate": 1.777416700907338e-05, | |
| "loss": 0.1656, | |
| "num_input_tokens_seen": 21471248, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 10.6875, | |
| "learning_rate": 1.623869018208499e-05, | |
| "loss": 0.1906, | |
| "num_input_tokens_seen": 22898704, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.53125, | |
| "grad_norm": 8.4375, | |
| "learning_rate": 1.4689996481586688e-05, | |
| "loss": 0.1883, | |
| "num_input_tokens_seen": 24365264, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5625, | |
| "grad_norm": 14.0, | |
| "learning_rate": 1.3144610530959784e-05, | |
| "loss": 0.2115, | |
| "num_input_tokens_seen": 25870176, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.59375, | |
| "grad_norm": 16.25, | |
| "learning_rate": 1.1619021659762912e-05, | |
| "loss": 0.1644, | |
| "num_input_tokens_seen": 27346880, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 8.375, | |
| "learning_rate": 1.0129507961929749e-05, | |
| "loss": 0.2046, | |
| "num_input_tokens_seen": 28825968, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.65625, | |
| "grad_norm": 13.3125, | |
| "learning_rate": 8.691962607859386e-06, | |
| "loss": 0.1767, | |
| "num_input_tokens_seen": 30278096, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6875, | |
| "grad_norm": 6.96875, | |
| "learning_rate": 7.321724263655989e-06, | |
| "loss": 0.1739, | |
| "num_input_tokens_seen": 31701248, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.71875, | |
| "grad_norm": 3.765625, | |
| "learning_rate": 6.0334134269513865e-06, | |
| "loss": 0.1648, | |
| "num_input_tokens_seen": 33175280, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 37.5, | |
| "learning_rate": 4.840776425613887e-06, | |
| "loss": 0.1662, | |
| "num_input_tokens_seen": 34680976, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.78125, | |
| "grad_norm": 6.3125, | |
| "learning_rate": 3.756538743883111e-06, | |
| "loss": 0.1867, | |
| "num_input_tokens_seen": 36137360, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8125, | |
| "grad_norm": 3.46875, | |
| "learning_rate": 2.792269240947076e-06, | |
| "loss": 0.1601, | |
| "num_input_tokens_seen": 37577552, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.84375, | |
| "grad_norm": 6.0, | |
| "learning_rate": 1.958256710754496e-06, | |
| "loss": 0.1679, | |
| "num_input_tokens_seen": 39032832, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 8.125, | |
| "learning_rate": 1.2634001001741375e-06, | |
| "loss": 0.1518, | |
| "num_input_tokens_seen": 40447376, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.90625, | |
| "grad_norm": 5.96875, | |
| "learning_rate": 7.151135568777839e-07, | |
| "loss": 0.171, | |
| "num_input_tokens_seen": 41890144, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 17.5, | |
| "learning_rate": 3.192473200896828e-07, | |
| "loss": 0.1702, | |
| "num_input_tokens_seen": 43291456, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.96875, | |
| "grad_norm": 20.5, | |
| "learning_rate": 8.002529830136163e-08, | |
| "loss": 0.1595, | |
| "num_input_tokens_seen": 44726672, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 13.3125, | |
| "learning_rate": 0.0, | |
| "loss": 0.1722, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cnn_balanced_accuracy": 0.5759533564450047, | |
| "eval_cnn_f1_score": 0.2608695652173913, | |
| "eval_cnn_loss": 0.31287845969200134, | |
| "eval_cnn_pr_auc": 0.42626378805395937, | |
| "eval_cnn_roc_auc": 0.7679027909094093, | |
| "eval_cnn_runtime": 10.722, | |
| "eval_cnn_samples_per_second": 52.043, | |
| "eval_cnn_steps_per_second": 0.839, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_xsum_balanced_accuracy": 0.749277038750723, | |
| "eval_xsum_f1_score": 0.7258382642998027, | |
| "eval_xsum_loss": 0.5868125557899475, | |
| "eval_xsum_pr_auc": 0.7976212402287128, | |
| "eval_xsum_roc_auc": 0.8217916586337639, | |
| "eval_xsum_runtime": 9.7581, | |
| "eval_xsum_samples_per_second": 57.183, | |
| "eval_xsum_steps_per_second": 0.922, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_medias_balanced_accuracy": 0.7074867769288893, | |
| "eval_medias_f1_score": 0.5704225352112676, | |
| "eval_medias_loss": 0.45626452565193176, | |
| "eval_medias_pr_auc": 0.6584352779944264, | |
| "eval_medias_roc_auc": 0.7960236336159854, | |
| "eval_medias_runtime": 9.9504, | |
| "eval_medias_samples_per_second": 72.962, | |
| "eval_medias_steps_per_second": 1.206, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_meetb_balanced_accuracy": 0.7702572347266881, | |
| "eval_meetb_f1_score": 0.6498194945848376, | |
| "eval_meetb_loss": 0.316259503364563, | |
| "eval_meetb_pr_auc": 0.7402648937803408, | |
| "eval_meetb_roc_auc": 0.8713504823151126, | |
| "eval_meetb_runtime": 10.6662, | |
| "eval_meetb_samples_per_second": 72.378, | |
| "eval_meetb_steps_per_second": 1.219, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_wice_balanced_accuracy": 0.8154247364773681, | |
| "eval_wice_f1_score": 0.8679245283018868, | |
| "eval_wice_loss": 0.4058681130409241, | |
| "eval_wice_pr_auc": 0.9493679290266217, | |
| "eval_wice_roc_auc": 0.8869679395995185, | |
| "eval_wice_runtime": 123.4927, | |
| "eval_wice_samples_per_second": 2.899, | |
| "eval_wice_steps_per_second": 0.049, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_reveal_balanced_accuracy": 0.8941412213740458, | |
| "eval_reveal_f1_score": 0.9196141479099679, | |
| "eval_reveal_loss": 0.2713985741138458, | |
| "eval_reveal_pr_auc": 0.9884053490966663, | |
| "eval_reveal_roc_auc": 0.960442748091603, | |
| "eval_reveal_runtime": 7.8148, | |
| "eval_reveal_samples_per_second": 218.815, | |
| "eval_reveal_steps_per_second": 3.455, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_claim_verify_balanced_accuracy": 0.7654793545023335, | |
| "eval_claim_verify_f1_score": 0.6621160409556314, | |
| "eval_claim_verify_loss": 0.47898775339126587, | |
| "eval_claim_verify_pr_auc": 0.6755500302254687, | |
| "eval_claim_verify_roc_auc": 0.8519738375912951, | |
| "eval_claim_verify_runtime": 261.7647, | |
| "eval_claim_verify_samples_per_second": 4.156, | |
| "eval_claim_verify_steps_per_second": 0.065, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_fact_check_balanced_accuracy": 0.7718129805113534, | |
| "eval_fact_check_f1_score": 0.9008264462809917, | |
| "eval_fact_check_loss": 0.39186760783195496, | |
| "eval_fact_check_pr_auc": 0.9477958353480682, | |
| "eval_fact_check_roc_auc": 0.8752134364384051, | |
| "eval_fact_check_runtime": 5.283, | |
| "eval_fact_check_samples_per_second": 296.421, | |
| "eval_fact_check_steps_per_second": 4.732, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_expertqa_balanced_accuracy": 0.5946286054753636, | |
| "eval_expertqa_f1_score": 0.36883320281910725, | |
| "eval_expertqa_loss": 1.1040500402450562, | |
| "eval_expertqa_pr_auc": 0.2890662808738864, | |
| "eval_expertqa_roc_auc": 0.6258950520788967, | |
| "eval_expertqa_runtime": 536.4175, | |
| "eval_expertqa_samples_per_second": 6.901, | |
| "eval_expertqa_steps_per_second": 0.108, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_lfqa_balanced_accuracy": 0.8775737079235312, | |
| "eval_lfqa_f1_score": 0.8557275541795665, | |
| "eval_lfqa_loss": 0.27868813276290894, | |
| "eval_lfqa_pr_auc": 0.9362703923849833, | |
| "eval_lfqa_roc_auc": 0.9523769464424847, | |
| "eval_lfqa_runtime": 12.7663, | |
| "eval_lfqa_samples_per_second": 149.691, | |
| "eval_lfqa_steps_per_second": 2.35, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_ragtruth_balanced_accuracy": 0.7859350741200968, | |
| "eval_ragtruth_f1_score": 0.5078776645041705, | |
| "eval_ragtruth_loss": 0.26025664806365967, | |
| "eval_ragtruth_pr_auc": 0.4400226298715947, | |
| "eval_ragtruth_roc_auc": 0.8972721245465168, | |
| "eval_ragtruth_runtime": 149.8477, | |
| "eval_ragtruth_samples_per_second": 109.251, | |
| "eval_ragtruth_steps_per_second": 1.708, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_halloumi_synthetic_balanced_accuracy": 0.7681937112954458, | |
| "eval_halloumi_synthetic_f1_score": 0.6775431861804223, | |
| "eval_halloumi_synthetic_loss": 0.34176450967788696, | |
| "eval_halloumi_synthetic_pr_auc": 0.7894696954251237, | |
| "eval_halloumi_synthetic_roc_auc": 0.9044560224857751, | |
| "eval_halloumi_synthetic_runtime": 38.8927, | |
| "eval_halloumi_synthetic_samples_per_second": 53.712, | |
| "eval_halloumi_synthetic_steps_per_second": 0.848, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "num_input_tokens_seen": 46209888, | |
| "step": 320, | |
| "total_flos": 2.418947678713938e+17, | |
| "train_loss": 0.2418923556804657, | |
| "train_runtime": 3758.6466, | |
| "train_samples_per_second": 10.887, | |
| "train_steps_per_second": 0.085, | |
| "train_tokens_per_second": 1545.744 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 320, | |
| "num_input_tokens_seen": 46209888, | |
| "num_train_epochs": 1, | |
| "save_steps": 0, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.418947678713938e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |