| { | |
| "best_metric": 2.1908211708068848, | |
| "best_model_checkpoint": "output/mistral_squad/checkpoint-1000", | |
| "epoch": 0.4910385465259023, | |
| "eval_steps": 1000, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 2.5e-05, | |
| "loss": 2.1039, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9227, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 1.8559, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0001, | |
| "loss": 1.8136, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.937075257991442e-05, | |
| "loss": 1.795, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.874150515982885e-05, | |
| "loss": 1.7844, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.811225773974327e-05, | |
| "loss": 1.7486, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.74830103196577e-05, | |
| "loss": 1.719, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.685376289957212e-05, | |
| "loss": 1.7017, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.622451547948655e-05, | |
| "loss": 1.6495, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.559526805940097e-05, | |
| "loss": 1.5982, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.496602063931538e-05, | |
| "loss": 1.5912, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.43367732192298e-05, | |
| "loss": 1.5603, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.370752579914423e-05, | |
| "loss": 1.5453, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.307827837905865e-05, | |
| "loss": 1.4757, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.244903095897306e-05, | |
| "loss": 1.4421, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.18197835388875e-05, | |
| "loss": 1.3738, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 9.119053611880191e-05, | |
| "loss": 1.3607, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.056128869871634e-05, | |
| "loss": 1.354, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 8.993204127863076e-05, | |
| "loss": 1.3505, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 8.930279385854519e-05, | |
| "loss": 1.3025, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 8.867354643845961e-05, | |
| "loss": 1.2577, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 8.804429901837403e-05, | |
| "loss": 1.3072, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.741505159828846e-05, | |
| "loss": 1.2621, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.678580417820287e-05, | |
| "loss": 1.213, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 8.615655675811729e-05, | |
| "loss": 1.1523, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.552730933803171e-05, | |
| "loss": 1.1324, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.489806191794614e-05, | |
| "loss": 1.1531, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 8.426881449786056e-05, | |
| "loss": 1.0632, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 8.363956707777499e-05, | |
| "loss": 1.1238, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 8.30103196576894e-05, | |
| "loss": 1.1197, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.238107223760383e-05, | |
| "loss": 1.04, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.175182481751825e-05, | |
| "loss": 1.1003, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8.112257739743267e-05, | |
| "loss": 1.0161, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8.04933299773471e-05, | |
| "loss": 0.972, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 7.986408255726152e-05, | |
| "loss": 0.9713, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 7.923483513717595e-05, | |
| "loss": 0.9788, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 7.860558771709036e-05, | |
| "loss": 0.9189, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 7.79763402970048e-05, | |
| "loss": 0.9001, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.734709287691921e-05, | |
| "loss": 0.8809, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 2.1908211708068848, | |
| "eval_runtime": 596.3037, | |
| "eval_samples_per_second": 19.911, | |
| "eval_steps_per_second": 0.624, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.671784545683363e-05, | |
| "loss": 0.8418, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 7.608859803674805e-05, | |
| "loss": 0.8633, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 7.545935061666248e-05, | |
| "loss": 0.8767, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 7.48301031965769e-05, | |
| "loss": 0.8018, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.420085577649131e-05, | |
| "loss": 0.817, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.357160835640574e-05, | |
| "loss": 0.7738, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.294236093632016e-05, | |
| "loss": 0.7264, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.231311351623459e-05, | |
| "loss": 0.8269, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.168386609614901e-05, | |
| "loss": 0.7176, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 7.105461867606344e-05, | |
| "loss": 0.7344, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 7.042537125597786e-05, | |
| "loss": 0.7132, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 6.979612383589229e-05, | |
| "loss": 0.6838, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 6.91668764158067e-05, | |
| "loss": 0.7343, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 6.853762899572112e-05, | |
| "loss": 0.6807, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 6.790838157563554e-05, | |
| "loss": 0.6712, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 6.727913415554995e-05, | |
| "loss": 0.6808, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 6.664988673546439e-05, | |
| "loss": 0.6305, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 6.60206393153788e-05, | |
| "loss": 0.645, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 6.539139189529323e-05, | |
| "loss": 0.6229, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 6.476214447520765e-05, | |
| "loss": 0.5976, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 6.413289705512208e-05, | |
| "loss": 0.6214, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.35036496350365e-05, | |
| "loss": 0.5981, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.287440221495093e-05, | |
| "loss": 0.564, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.224515479486535e-05, | |
| "loss": 0.5757, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 6.161590737477976e-05, | |
| "loss": 0.5611, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 6.0986659954694194e-05, | |
| "loss": 0.5462, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 6.035741253460862e-05, | |
| "loss": 0.5172, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 5.9728165114523035e-05, | |
| "loss": 0.5087, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 5.909891769443746e-05, | |
| "loss": 0.5148, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 5.8469670274351876e-05, | |
| "loss": 0.5247, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 5.784042285426629e-05, | |
| "loss": 0.5356, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 5.721117543418072e-05, | |
| "loss": 0.4837, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 5.658192801409514e-05, | |
| "loss": 0.533, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 5.5952680594009565e-05, | |
| "loss": 0.485, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 5.532343317392399e-05, | |
| "loss": 0.4728, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 5.469418575383841e-05, | |
| "loss": 0.4873, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 5.406493833375284e-05, | |
| "loss": 0.4693, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 5.343569091366726e-05, | |
| "loss": 0.4951, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 5.280644349358168e-05, | |
| "loss": 0.4445, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 5.21771960734961e-05, | |
| "loss": 0.4423, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_loss": 2.3274576663970947, | |
| "eval_runtime": 596.4629, | |
| "eval_samples_per_second": 19.906, | |
| "eval_steps_per_second": 0.624, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 4073, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1000, | |
| "total_flos": 7.64977027743744e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |