{ "best_metric": null, "best_model_checkpoint": null, "epoch": 4.0, "global_step": 89992, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.02, "learning_rate": 4.972219752866922e-05, "loss": 2.1683, "step": 500 }, { "epoch": 0.04, "learning_rate": 4.9444395057338435e-05, "loss": 1.892, "step": 1000 }, { "epoch": 0.07, "learning_rate": 4.9166592586007645e-05, "loss": 1.8415, "step": 1500 }, { "epoch": 0.09, "learning_rate": 4.888879011467686e-05, "loss": 1.7915, "step": 2000 }, { "epoch": 0.11, "learning_rate": 4.861098764334608e-05, "loss": 1.7884, "step": 2500 }, { "epoch": 0.13, "learning_rate": 4.8333185172015294e-05, "loss": 1.7897, "step": 3000 }, { "epoch": 0.16, "learning_rate": 4.805538270068451e-05, "loss": 1.7978, "step": 3500 }, { "epoch": 0.18, "learning_rate": 4.777758022935373e-05, "loss": 1.7701, "step": 4000 }, { "epoch": 0.2, "learning_rate": 4.7499777758022936e-05, "loss": 1.7776, "step": 4500 }, { "epoch": 0.22, "learning_rate": 4.722197528669215e-05, "loss": 1.7768, "step": 5000 }, { "epoch": 0.24, "learning_rate": 4.694417281536137e-05, "loss": 1.784, "step": 5500 }, { "epoch": 0.27, "learning_rate": 4.6666370344030586e-05, "loss": 1.7417, "step": 6000 }, { "epoch": 0.29, "learning_rate": 4.6388567872699795e-05, "loss": 1.7433, "step": 6500 }, { "epoch": 0.31, "learning_rate": 4.611076540136901e-05, "loss": 1.7252, "step": 7000 }, { "epoch": 0.33, "learning_rate": 4.583296293003823e-05, "loss": 1.7322, "step": 7500 }, { "epoch": 0.36, "learning_rate": 4.555516045870744e-05, "loss": 1.7337, "step": 8000 }, { "epoch": 0.38, "learning_rate": 4.5277357987376654e-05, "loss": 1.7481, "step": 8500 }, { "epoch": 0.4, "learning_rate": 4.499955551604587e-05, "loss": 1.7261, "step": 9000 }, { "epoch": 0.42, "learning_rate": 4.472175304471509e-05, "loss": 1.7654, "step": 9500 }, { "epoch": 0.44, "learning_rate": 4.44439505733843e-05, "loss": 1.7241, "step": 10000 }, { "epoch": 0.47, "learning_rate": 4.416614810205352e-05, "loss": 1.7256, "step": 10500 }, { "epoch": 0.49, "learning_rate": 4.388834563072273e-05, "loss": 1.7115, "step": 11000 }, { "epoch": 0.51, "learning_rate": 4.3610543159391946e-05, "loss": 1.7019, "step": 11500 }, { "epoch": 0.53, "learning_rate": 4.333274068806116e-05, "loss": 1.7066, "step": 12000 }, { "epoch": 0.56, "learning_rate": 4.305493821673038e-05, "loss": 1.7005, "step": 12500 }, { "epoch": 0.58, "learning_rate": 4.2777135745399595e-05, "loss": 1.7056, "step": 13000 }, { "epoch": 0.6, "learning_rate": 4.249933327406881e-05, "loss": 1.7052, "step": 13500 }, { "epoch": 0.62, "learning_rate": 4.222153080273802e-05, "loss": 1.7141, "step": 14000 }, { "epoch": 0.64, "learning_rate": 4.194372833140724e-05, "loss": 1.7128, "step": 14500 }, { "epoch": 0.67, "learning_rate": 4.1665925860076454e-05, "loss": 1.6943, "step": 15000 }, { "epoch": 0.69, "learning_rate": 4.138812338874567e-05, "loss": 1.7187, "step": 15500 }, { "epoch": 0.71, "learning_rate": 4.1110320917414887e-05, "loss": 1.7134, "step": 16000 }, { "epoch": 0.73, "learning_rate": 4.08325184460841e-05, "loss": 1.7088, "step": 16500 }, { "epoch": 0.76, "learning_rate": 4.055471597475331e-05, "loss": 1.6984, "step": 17000 }, { "epoch": 0.78, "learning_rate": 4.027691350342253e-05, "loss": 1.7107, "step": 17500 }, { "epoch": 0.8, "learning_rate": 3.9999111032091745e-05, "loss": 1.7043, "step": 18000 }, { "epoch": 0.82, "learning_rate": 3.9721308560760955e-05, "loss": 1.7023, "step": 18500 }, { "epoch": 0.84, "learning_rate": 3.944350608943017e-05, "loss": 1.6946, "step": 19000 }, { "epoch": 0.87, "learning_rate": 3.916570361809939e-05, "loss": 1.6791, "step": 19500 }, { "epoch": 0.89, "learning_rate": 3.8887901146768604e-05, "loss": 1.7051, "step": 20000 }, { "epoch": 0.91, "learning_rate": 3.8610098675437814e-05, "loss": 1.6723, "step": 20500 }, { "epoch": 0.93, "learning_rate": 3.833229620410703e-05, "loss": 1.7001, "step": 21000 }, { "epoch": 0.96, "learning_rate": 3.8054493732776247e-05, "loss": 1.6728, "step": 21500 }, { "epoch": 0.98, "learning_rate": 3.777669126144546e-05, "loss": 1.6689, "step": 22000 }, { "epoch": 1.0, "learning_rate": 3.749888879011468e-05, "loss": 1.6921, "step": 22500 }, { "epoch": 1.02, "learning_rate": 3.7221086318783896e-05, "loss": 1.653, "step": 23000 }, { "epoch": 1.04, "learning_rate": 3.6943283847453105e-05, "loss": 1.6692, "step": 23500 }, { "epoch": 1.07, "learning_rate": 3.666548137612232e-05, "loss": 1.6404, "step": 24000 }, { "epoch": 1.09, "learning_rate": 3.638767890479154e-05, "loss": 1.6781, "step": 24500 }, { "epoch": 1.11, "learning_rate": 3.6109876433460755e-05, "loss": 1.6622, "step": 25000 }, { "epoch": 1.13, "learning_rate": 3.583207396212997e-05, "loss": 1.645, "step": 25500 }, { "epoch": 1.16, "learning_rate": 3.555427149079919e-05, "loss": 1.6476, "step": 26000 }, { "epoch": 1.18, "learning_rate": 3.52764690194684e-05, "loss": 1.6429, "step": 26500 }, { "epoch": 1.2, "learning_rate": 3.4998666548137613e-05, "loss": 1.6649, "step": 27000 }, { "epoch": 1.22, "learning_rate": 3.472086407680683e-05, "loss": 1.6437, "step": 27500 }, { "epoch": 1.24, "learning_rate": 3.4443061605476046e-05, "loss": 1.6383, "step": 28000 }, { "epoch": 1.27, "learning_rate": 3.416525913414526e-05, "loss": 1.6574, "step": 28500 }, { "epoch": 1.29, "learning_rate": 3.388745666281448e-05, "loss": 1.6443, "step": 29000 }, { "epoch": 1.31, "learning_rate": 3.360965419148369e-05, "loss": 1.6584, "step": 29500 }, { "epoch": 1.33, "learning_rate": 3.3331851720152905e-05, "loss": 1.6487, "step": 30000 }, { "epoch": 1.36, "learning_rate": 3.3054049248822115e-05, "loss": 1.6691, "step": 30500 }, { "epoch": 1.38, "learning_rate": 3.277624677749133e-05, "loss": 1.663, "step": 31000 }, { "epoch": 1.4, "learning_rate": 3.249844430616055e-05, "loss": 1.6332, "step": 31500 }, { "epoch": 1.42, "learning_rate": 3.2220641834829764e-05, "loss": 1.6474, "step": 32000 }, { "epoch": 1.44, "learning_rate": 3.194283936349898e-05, "loss": 1.6257, "step": 32500 }, { "epoch": 1.47, "learning_rate": 3.166503689216819e-05, "loss": 1.6491, "step": 33000 }, { "epoch": 1.49, "learning_rate": 3.1387234420837406e-05, "loss": 1.6215, "step": 33500 }, { "epoch": 1.51, "learning_rate": 3.110943194950662e-05, "loss": 1.6691, "step": 34000 }, { "epoch": 1.53, "learning_rate": 3.083162947817584e-05, "loss": 1.6339, "step": 34500 }, { "epoch": 1.56, "learning_rate": 3.0553827006845055e-05, "loss": 1.6701, "step": 35000 }, { "epoch": 1.58, "learning_rate": 3.027602453551427e-05, "loss": 1.6105, "step": 35500 }, { "epoch": 1.6, "learning_rate": 2.9998222064183485e-05, "loss": 1.6545, "step": 36000 }, { "epoch": 1.62, "learning_rate": 2.97204195928527e-05, "loss": 1.6452, "step": 36500 }, { "epoch": 1.64, "learning_rate": 2.9442617121521914e-05, "loss": 1.6699, "step": 37000 }, { "epoch": 1.67, "learning_rate": 2.916481465019113e-05, "loss": 1.6366, "step": 37500 }, { "epoch": 1.69, "learning_rate": 2.8887012178860344e-05, "loss": 1.624, "step": 38000 }, { "epoch": 1.71, "learning_rate": 2.860920970752956e-05, "loss": 1.6441, "step": 38500 }, { "epoch": 1.73, "learning_rate": 2.8331407236198777e-05, "loss": 1.6528, "step": 39000 }, { "epoch": 1.76, "learning_rate": 2.805360476486799e-05, "loss": 1.6217, "step": 39500 }, { "epoch": 1.78, "learning_rate": 2.7775802293537206e-05, "loss": 1.6429, "step": 40000 }, { "epoch": 1.8, "learning_rate": 2.7497999822206422e-05, "loss": 1.6334, "step": 40500 }, { "epoch": 1.82, "learning_rate": 2.7220197350875635e-05, "loss": 1.6595, "step": 41000 }, { "epoch": 1.84, "learning_rate": 2.6942394879544852e-05, "loss": 1.636, "step": 41500 }, { "epoch": 1.87, "learning_rate": 2.6664592408214068e-05, "loss": 1.6176, "step": 42000 }, { "epoch": 1.89, "learning_rate": 2.6386789936883278e-05, "loss": 1.6374, "step": 42500 }, { "epoch": 1.91, "learning_rate": 2.610898746555249e-05, "loss": 1.6541, "step": 43000 }, { "epoch": 1.93, "learning_rate": 2.5831184994221707e-05, "loss": 1.6518, "step": 43500 }, { "epoch": 1.96, "learning_rate": 2.5553382522890924e-05, "loss": 1.645, "step": 44000 }, { "epoch": 1.98, "learning_rate": 2.5275580051560137e-05, "loss": 1.637, "step": 44500 }, { "epoch": 2.0, "learning_rate": 2.4997777580229356e-05, "loss": 1.6229, "step": 45000 }, { "epoch": 2.02, "learning_rate": 2.471997510889857e-05, "loss": 1.601, "step": 45500 }, { "epoch": 2.04, "learning_rate": 2.4442172637567782e-05, "loss": 1.6071, "step": 46000 }, { "epoch": 2.07, "learning_rate": 2.4164370166237e-05, "loss": 1.6028, "step": 46500 }, { "epoch": 2.09, "learning_rate": 2.3886567694906215e-05, "loss": 1.6091, "step": 47000 }, { "epoch": 2.11, "learning_rate": 2.3608765223575428e-05, "loss": 1.5878, "step": 47500 }, { "epoch": 2.13, "learning_rate": 2.3330962752244645e-05, "loss": 1.6169, "step": 48000 }, { "epoch": 2.16, "learning_rate": 2.305316028091386e-05, "loss": 1.6102, "step": 48500 }, { "epoch": 2.18, "learning_rate": 2.2775357809583074e-05, "loss": 1.6253, "step": 49000 }, { "epoch": 2.2, "learning_rate": 2.249755533825229e-05, "loss": 1.636, "step": 49500 }, { "epoch": 2.22, "learning_rate": 2.2219752866921507e-05, "loss": 1.6303, "step": 50000 }, { "epoch": 2.24, "learning_rate": 2.194195039559072e-05, "loss": 1.6342, "step": 50500 }, { "epoch": 2.27, "learning_rate": 2.1664147924259936e-05, "loss": 1.6082, "step": 51000 }, { "epoch": 2.29, "learning_rate": 2.138634545292915e-05, "loss": 1.6241, "step": 51500 }, { "epoch": 2.31, "learning_rate": 2.1108542981598366e-05, "loss": 1.6364, "step": 52000 }, { "epoch": 2.33, "learning_rate": 2.083074051026758e-05, "loss": 1.5943, "step": 52500 }, { "epoch": 2.36, "learning_rate": 2.0552938038936795e-05, "loss": 1.5904, "step": 53000 }, { "epoch": 2.38, "learning_rate": 2.027513556760601e-05, "loss": 1.6107, "step": 53500 }, { "epoch": 2.4, "learning_rate": 1.9997333096275224e-05, "loss": 1.623, "step": 54000 }, { "epoch": 2.42, "learning_rate": 1.971953062494444e-05, "loss": 1.5865, "step": 54500 }, { "epoch": 2.44, "learning_rate": 1.9441728153613657e-05, "loss": 1.6293, "step": 55000 }, { "epoch": 2.47, "learning_rate": 1.916392568228287e-05, "loss": 1.6075, "step": 55500 }, { "epoch": 2.49, "learning_rate": 1.8886123210952087e-05, "loss": 1.6136, "step": 56000 }, { "epoch": 2.51, "learning_rate": 1.8608320739621303e-05, "loss": 1.6275, "step": 56500 }, { "epoch": 2.53, "learning_rate": 1.8330518268290516e-05, "loss": 1.604, "step": 57000 }, { "epoch": 2.56, "learning_rate": 1.805271579695973e-05, "loss": 1.6248, "step": 57500 }, { "epoch": 2.58, "learning_rate": 1.7774913325628945e-05, "loss": 1.6077, "step": 58000 }, { "epoch": 2.6, "learning_rate": 1.749711085429816e-05, "loss": 1.6032, "step": 58500 }, { "epoch": 2.62, "learning_rate": 1.7219308382967375e-05, "loss": 1.6181, "step": 59000 }, { "epoch": 2.64, "learning_rate": 1.694150591163659e-05, "loss": 1.6061, "step": 59500 }, { "epoch": 2.67, "learning_rate": 1.6663703440305804e-05, "loss": 1.5769, "step": 60000 }, { "epoch": 2.69, "learning_rate": 1.638590096897502e-05, "loss": 1.6039, "step": 60500 }, { "epoch": 2.71, "learning_rate": 1.6108098497644237e-05, "loss": 1.6071, "step": 61000 }, { "epoch": 2.73, "learning_rate": 1.583029602631345e-05, "loss": 1.5944, "step": 61500 }, { "epoch": 2.76, "learning_rate": 1.5552493554982667e-05, "loss": 1.5803, "step": 62000 }, { "epoch": 2.78, "learning_rate": 1.5274691083651883e-05, "loss": 1.6037, "step": 62500 }, { "epoch": 2.8, "learning_rate": 1.4996888612321098e-05, "loss": 1.5921, "step": 63000 }, { "epoch": 2.82, "learning_rate": 1.4719086140990309e-05, "loss": 1.6005, "step": 63500 }, { "epoch": 2.84, "learning_rate": 1.4441283669659525e-05, "loss": 1.5907, "step": 64000 }, { "epoch": 2.87, "learning_rate": 1.416348119832874e-05, "loss": 1.6009, "step": 64500 }, { "epoch": 2.89, "learning_rate": 1.3885678726997955e-05, "loss": 1.5981, "step": 65000 }, { "epoch": 2.91, "learning_rate": 1.3607876255667171e-05, "loss": 1.6182, "step": 65500 }, { "epoch": 2.93, "learning_rate": 1.3330073784336386e-05, "loss": 1.6073, "step": 66000 }, { "epoch": 2.96, "learning_rate": 1.30522713130056e-05, "loss": 1.6208, "step": 66500 }, { "epoch": 2.98, "learning_rate": 1.2774468841674817e-05, "loss": 1.5967, "step": 67000 }, { "epoch": 3.0, "learning_rate": 1.2496666370344032e-05, "loss": 1.6261, "step": 67500 }, { "epoch": 3.02, "learning_rate": 1.2218863899013246e-05, "loss": 1.5709, "step": 68000 }, { "epoch": 3.04, "learning_rate": 1.1941061427682461e-05, "loss": 1.604, "step": 68500 }, { "epoch": 3.07, "learning_rate": 1.1663258956351676e-05, "loss": 1.5782, "step": 69000 }, { "epoch": 3.09, "learning_rate": 1.1385456485020892e-05, "loss": 1.5971, "step": 69500 }, { "epoch": 3.11, "learning_rate": 1.1107654013690107e-05, "loss": 1.6089, "step": 70000 }, { "epoch": 3.13, "learning_rate": 1.0829851542359322e-05, "loss": 1.5871, "step": 70500 }, { "epoch": 3.16, "learning_rate": 1.0552049071028536e-05, "loss": 1.5854, "step": 71000 }, { "epoch": 3.18, "learning_rate": 1.0274246599697751e-05, "loss": 1.6105, "step": 71500 }, { "epoch": 3.2, "learning_rate": 9.996444128366966e-06, "loss": 1.5779, "step": 72000 }, { "epoch": 3.22, "learning_rate": 9.718641657036182e-06, "loss": 1.5993, "step": 72500 }, { "epoch": 3.24, "learning_rate": 9.440839185705397e-06, "loss": 1.5811, "step": 73000 }, { "epoch": 3.27, "learning_rate": 9.163036714374612e-06, "loss": 1.5707, "step": 73500 }, { "epoch": 3.29, "learning_rate": 8.885234243043826e-06, "loss": 1.5867, "step": 74000 }, { "epoch": 3.31, "learning_rate": 8.607431771713041e-06, "loss": 1.6162, "step": 74500 }, { "epoch": 3.33, "learning_rate": 8.329629300382257e-06, "loss": 1.6259, "step": 75000 }, { "epoch": 3.36, "learning_rate": 8.051826829051472e-06, "loss": 1.5916, "step": 75500 }, { "epoch": 3.38, "learning_rate": 7.774024357720687e-06, "loss": 1.5834, "step": 76000 }, { "epoch": 3.4, "learning_rate": 7.496221886389902e-06, "loss": 1.593, "step": 76500 }, { "epoch": 3.42, "learning_rate": 7.218419415059116e-06, "loss": 1.5937, "step": 77000 }, { "epoch": 3.44, "learning_rate": 6.940616943728332e-06, "loss": 1.5642, "step": 77500 }, { "epoch": 3.47, "learning_rate": 6.662814472397546e-06, "loss": 1.575, "step": 78000 }, { "epoch": 3.49, "learning_rate": 6.385012001066762e-06, "loss": 1.5726, "step": 78500 }, { "epoch": 3.51, "learning_rate": 6.1072095297359775e-06, "loss": 1.5616, "step": 79000 }, { "epoch": 3.53, "learning_rate": 5.829407058405191e-06, "loss": 1.5951, "step": 79500 }, { "epoch": 3.56, "learning_rate": 5.551604587074407e-06, "loss": 1.6033, "step": 80000 }, { "epoch": 3.58, "learning_rate": 5.2738021157436225e-06, "loss": 1.593, "step": 80500 }, { "epoch": 3.6, "learning_rate": 4.995999644412837e-06, "loss": 1.5798, "step": 81000 }, { "epoch": 3.62, "learning_rate": 4.718197173082052e-06, "loss": 1.5774, "step": 81500 }, { "epoch": 3.64, "learning_rate": 4.4403947017512674e-06, "loss": 1.5894, "step": 82000 }, { "epoch": 3.67, "learning_rate": 4.162592230420482e-06, "loss": 1.5609, "step": 82500 }, { "epoch": 3.69, "learning_rate": 3.884789759089697e-06, "loss": 1.5854, "step": 83000 }, { "epoch": 3.71, "learning_rate": 3.6069872877589124e-06, "loss": 1.5871, "step": 83500 }, { "epoch": 3.73, "learning_rate": 3.329184816428127e-06, "loss": 1.5745, "step": 84000 }, { "epoch": 3.76, "learning_rate": 3.0513823450973422e-06, "loss": 1.5986, "step": 84500 }, { "epoch": 3.78, "learning_rate": 2.773579873766557e-06, "loss": 1.5986, "step": 85000 }, { "epoch": 3.8, "learning_rate": 2.495777402435772e-06, "loss": 1.5965, "step": 85500 }, { "epoch": 3.82, "learning_rate": 2.217974931104987e-06, "loss": 1.583, "step": 86000 }, { "epoch": 3.84, "learning_rate": 1.9401724597742023e-06, "loss": 1.5983, "step": 86500 }, { "epoch": 3.87, "learning_rate": 1.662369988443417e-06, "loss": 1.5991, "step": 87000 }, { "epoch": 3.89, "learning_rate": 1.3845675171126324e-06, "loss": 1.5761, "step": 87500 }, { "epoch": 3.91, "learning_rate": 1.1067650457818473e-06, "loss": 1.583, "step": 88000 }, { "epoch": 3.93, "learning_rate": 8.289625744510624e-07, "loss": 1.6017, "step": 88500 }, { "epoch": 3.96, "learning_rate": 5.511601031202774e-07, "loss": 1.574, "step": 89000 }, { "epoch": 3.98, "learning_rate": 2.733576317894924e-07, "loss": 1.5734, "step": 89500 }, { "epoch": 4.0, "step": 89992, "total_flos": 2.1371909200869427e+17, "train_loss": 1.6458884161285257, "train_runtime": 76560.3935, "train_samples_per_second": 3.526, "train_steps_per_second": 1.175 } ], "max_steps": 89992, "num_train_epochs": 4, "total_flos": 2.1371909200869427e+17, "trial_name": null, "trial_params": null }