joelniklaus's picture
Training in progress, step 150000
fddae7d
raw
history blame
19.2 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 96.00168,
"global_step": 150000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 1e-05,
"loss": 5.7318,
"step": 1000
},
{
"epoch": 1.0,
"learning_rate": 2e-05,
"loss": 3.9475,
"step": 2000
},
{
"epoch": 1.01,
"learning_rate": 3e-05,
"loss": 3.4547,
"step": 3000
},
{
"epoch": 2.0,
"learning_rate": 4e-05,
"loss": 2.9629,
"step": 4000
},
{
"epoch": 3.0,
"learning_rate": 5e-05,
"loss": 2.727,
"step": 5000
},
{
"epoch": 3.01,
"learning_rate": 6e-05,
"loss": 1.8697,
"step": 6000
},
{
"epoch": 4.0,
"learning_rate": 7e-05,
"loss": 1.5021,
"step": 7000
},
{
"epoch": 5.0,
"learning_rate": 8e-05,
"loss": 1.4643,
"step": 8000
},
{
"epoch": 5.01,
"learning_rate": 9e-05,
"loss": 1.2784,
"step": 9000
},
{
"epoch": 6.0,
"learning_rate": 0.0001,
"loss": 1.2059,
"step": 10000
},
{
"epoch": 7.0,
"learning_rate": 9.999316524962345e-05,
"loss": 1.2602,
"step": 11000
},
{
"epoch": 7.01,
"learning_rate": 9.997266286704631e-05,
"loss": 1.1214,
"step": 12000
},
{
"epoch": 8.0,
"learning_rate": 9.993849845741524e-05,
"loss": 1.0983,
"step": 13000
},
{
"epoch": 8.01,
"learning_rate": 9.989068136093873e-05,
"loss": 1.1488,
"step": 14000
},
{
"epoch": 9.0,
"learning_rate": 9.98292246503335e-05,
"loss": 1.0281,
"step": 15000
},
{
"epoch": 10.0,
"learning_rate": 9.975414512725057e-05,
"loss": 1.0586,
"step": 16000
},
{
"epoch": 10.01,
"learning_rate": 9.966546331768191e-05,
"loss": 1.062,
"step": 17000
},
{
"epoch": 11.0,
"learning_rate": 9.956320346634876e-05,
"loss": 0.973,
"step": 18000
},
{
"epoch": 12.0,
"learning_rate": 9.944739353007344e-05,
"loss": 1.0381,
"step": 19000
},
{
"epoch": 12.01,
"learning_rate": 9.931806517013612e-05,
"loss": 0.9823,
"step": 20000
},
{
"epoch": 13.0,
"learning_rate": 9.917525374361912e-05,
"loss": 0.9447,
"step": 21000
},
{
"epoch": 14.0,
"learning_rate": 9.901899829374047e-05,
"loss": 1.0206,
"step": 22000
},
{
"epoch": 14.01,
"learning_rate": 9.884934153917997e-05,
"loss": 0.9389,
"step": 23000
},
{
"epoch": 15.0,
"learning_rate": 9.86663298624003e-05,
"loss": 0.9147,
"step": 24000
},
{
"epoch": 16.0,
"learning_rate": 9.847001329696653e-05,
"loss": 1.0002,
"step": 25000
},
{
"epoch": 16.01,
"learning_rate": 9.826044551386744e-05,
"loss": 0.908,
"step": 26000
},
{
"epoch": 17.0,
"learning_rate": 9.803768380684242e-05,
"loss": 0.9088,
"step": 27000
},
{
"epoch": 17.01,
"learning_rate": 9.780178907671789e-05,
"loss": 0.9641,
"step": 28000
},
{
"epoch": 18.0,
"learning_rate": 9.755282581475769e-05,
"loss": 0.8728,
"step": 29000
},
{
"epoch": 19.0,
"learning_rate": 9.729086208503174e-05,
"loss": 0.9131,
"step": 30000
},
{
"epoch": 19.01,
"learning_rate": 9.701596950580806e-05,
"loss": 0.924,
"step": 31000
},
{
"epoch": 20.0,
"learning_rate": 9.672822322997305e-05,
"loss": 0.8533,
"step": 32000
},
{
"epoch": 21.0,
"learning_rate": 9.642770192448536e-05,
"loss": 0.9286,
"step": 33000
},
{
"epoch": 21.01,
"learning_rate": 9.611448774886924e-05,
"loss": 0.8686,
"step": 34000
},
{
"epoch": 22.0,
"learning_rate": 9.578866633275288e-05,
"loss": 0.8435,
"step": 35000
},
{
"epoch": 23.0,
"learning_rate": 9.545032675245813e-05,
"loss": 0.9246,
"step": 36000
},
{
"epoch": 23.01,
"learning_rate": 9.509956150664796e-05,
"loss": 0.8463,
"step": 37000
},
{
"epoch": 24.0,
"learning_rate": 9.473646649103818e-05,
"loss": 0.839,
"step": 38000
},
{
"epoch": 25.0,
"learning_rate": 9.43611409721806e-05,
"loss": 0.9091,
"step": 39000
},
{
"epoch": 25.01,
"learning_rate": 9.397368756032445e-05,
"loss": 0.8261,
"step": 40000
},
{
"epoch": 26.0,
"learning_rate": 9.357421218136386e-05,
"loss": 0.8504,
"step": 41000
},
{
"epoch": 26.01,
"learning_rate": 9.316282404787871e-05,
"loss": 0.879,
"step": 42000
},
{
"epoch": 27.0,
"learning_rate": 9.273963562927695e-05,
"loss": 0.8044,
"step": 43000
},
{
"epoch": 28.0,
"learning_rate": 9.230476262104677e-05,
"loss": 0.8566,
"step": 44000
},
{
"epoch": 28.01,
"learning_rate": 9.185832391312644e-05,
"loss": 0.8478,
"step": 45000
},
{
"epoch": 29.0,
"learning_rate": 9.140044155740101e-05,
"loss": 0.7946,
"step": 46000
},
{
"epoch": 30.0,
"learning_rate": 9.093124073433463e-05,
"loss": 0.8694,
"step": 47000
},
{
"epoch": 30.01,
"learning_rate": 9.045084971874738e-05,
"loss": 0.8127,
"step": 48000
},
{
"epoch": 31.0,
"learning_rate": 8.995939984474624e-05,
"loss": 0.7912,
"step": 49000
},
{
"epoch": 32.0,
"learning_rate": 8.945702546981969e-05,
"loss": 0.8722,
"step": 50000
},
{
"epoch": 32.0,
"eval_loss": 0.5036267042160034,
"eval_runtime": 16.6411,
"eval_samples_per_second": 300.462,
"eval_steps_per_second": 2.404,
"step": 50000
},
{
"epoch": 32.01,
"learning_rate": 8.894386393810563e-05,
"loss": 0.7953,
"step": 51000
},
{
"epoch": 33.0,
"learning_rate": 8.842005554284296e-05,
"loss": 0.7928,
"step": 52000
},
{
"epoch": 33.01,
"learning_rate": 8.788574348801675e-05,
"loss": 0.8643,
"step": 53000
},
{
"epoch": 34.0,
"learning_rate": 8.73410738492077e-05,
"loss": 0.7746,
"step": 54000
},
{
"epoch": 35.0,
"learning_rate": 8.678619553365659e-05,
"loss": 0.8094,
"step": 55000
},
{
"epoch": 35.01,
"learning_rate": 8.622126023955446e-05,
"loss": 0.8318,
"step": 56000
},
{
"epoch": 36.0,
"learning_rate": 8.564642241456986e-05,
"loss": 0.7601,
"step": 57000
},
{
"epoch": 37.0,
"learning_rate": 8.506183921362443e-05,
"loss": 0.8153,
"step": 58000
},
{
"epoch": 37.01,
"learning_rate": 8.44676704559283e-05,
"loss": 0.8049,
"step": 59000
},
{
"epoch": 38.0,
"learning_rate": 8.386407858128706e-05,
"loss": 0.758,
"step": 60000
},
{
"epoch": 39.0,
"learning_rate": 8.32512286056924e-05,
"loss": 0.8346,
"step": 61000
},
{
"epoch": 39.01,
"learning_rate": 8.262928807620843e-05,
"loss": 0.7726,
"step": 62000
},
{
"epoch": 40.0,
"learning_rate": 8.199842702516583e-05,
"loss": 0.754,
"step": 63000
},
{
"epoch": 41.0,
"learning_rate": 8.135881792367686e-05,
"loss": 0.8374,
"step": 64000
},
{
"epoch": 41.01,
"learning_rate": 8.07106356344834e-05,
"loss": 0.7596,
"step": 65000
},
{
"epoch": 42.0,
"learning_rate": 8.005405736415126e-05,
"loss": 0.76,
"step": 66000
},
{
"epoch": 42.01,
"learning_rate": 7.938926261462366e-05,
"loss": 0.8258,
"step": 67000
},
{
"epoch": 43.0,
"learning_rate": 7.871643313414718e-05,
"loss": 0.7399,
"step": 68000
},
{
"epoch": 44.0,
"learning_rate": 7.803575286758364e-05,
"loss": 0.7789,
"step": 69000
},
{
"epoch": 44.01,
"learning_rate": 7.734740790612136e-05,
"loss": 0.801,
"step": 70000
},
{
"epoch": 45.0,
"learning_rate": 7.66515864363997e-05,
"loss": 0.729,
"step": 71000
},
{
"epoch": 46.0,
"learning_rate": 7.594847868906076e-05,
"loss": 0.7974,
"step": 72000
},
{
"epoch": 46.01,
"learning_rate": 7.52382768867422e-05,
"loss": 0.7615,
"step": 73000
},
{
"epoch": 47.0,
"learning_rate": 7.452117519152542e-05,
"loss": 0.7302,
"step": 74000
},
{
"epoch": 48.0,
"learning_rate": 7.379736965185368e-05,
"loss": 0.8065,
"step": 75000
},
{
"epoch": 48.01,
"learning_rate": 7.30670581489344e-05,
"loss": 0.7451,
"step": 76000
},
{
"epoch": 49.0,
"learning_rate": 7.233044034264034e-05,
"loss": 0.7263,
"step": 77000
},
{
"epoch": 50.0,
"learning_rate": 7.158771761692464e-05,
"loss": 0.8105,
"step": 78000
},
{
"epoch": 50.01,
"learning_rate": 7.083909302476453e-05,
"loss": 0.7318,
"step": 79000
},
{
"epoch": 51.0,
"learning_rate": 7.008477123264848e-05,
"loss": 0.7394,
"step": 80000
},
{
"epoch": 51.01,
"learning_rate": 6.932495846462261e-05,
"loss": 0.7932,
"step": 81000
},
{
"epoch": 52.0,
"learning_rate": 6.855986244591104e-05,
"loss": 0.7127,
"step": 82000
},
{
"epoch": 53.0,
"learning_rate": 6.778969234612584e-05,
"loss": 0.7575,
"step": 83000
},
{
"epoch": 53.01,
"learning_rate": 6.701465872208216e-05,
"loss": 0.7684,
"step": 84000
},
{
"epoch": 54.0,
"learning_rate": 6.623497346023418e-05,
"loss": 0.7064,
"step": 85000
},
{
"epoch": 55.0,
"learning_rate": 6.545084971874738e-05,
"loss": 0.7796,
"step": 86000
},
{
"epoch": 55.01,
"learning_rate": 6.466250186922325e-05,
"loss": 0.7313,
"step": 87000
},
{
"epoch": 56.0,
"learning_rate": 6.387014543809223e-05,
"loss": 0.7073,
"step": 88000
},
{
"epoch": 57.0,
"learning_rate": 6.307399704769099e-05,
"loss": 0.7876,
"step": 89000
},
{
"epoch": 57.01,
"learning_rate": 6.227427435703997e-05,
"loss": 0.718,
"step": 90000
},
{
"epoch": 58.0,
"learning_rate": 6.147119600233758e-05,
"loss": 0.7123,
"step": 91000
},
{
"epoch": 59.0,
"learning_rate": 6.066498153718735e-05,
"loss": 0.7829,
"step": 92000
},
{
"epoch": 59.01,
"learning_rate": 5.985585137257401e-05,
"loss": 0.7043,
"step": 93000
},
{
"epoch": 60.0,
"learning_rate": 5.90440267166055e-05,
"loss": 0.7318,
"step": 94000
},
{
"epoch": 60.01,
"learning_rate": 5.8229729514036705e-05,
"loss": 0.7595,
"step": 95000
},
{
"epoch": 61.0,
"learning_rate": 5.74131823855921e-05,
"loss": 0.6896,
"step": 96000
},
{
"epoch": 62.0,
"learning_rate": 5.6594608567103456e-05,
"loss": 0.7431,
"step": 97000
},
{
"epoch": 62.01,
"learning_rate": 5.577423184847932e-05,
"loss": 0.7372,
"step": 98000
},
{
"epoch": 63.0,
"learning_rate": 5.495227651252315e-05,
"loss": 0.6896,
"step": 99000
},
{
"epoch": 64.0,
"learning_rate": 5.4128967273616625e-05,
"loss": 0.7612,
"step": 100000
},
{
"epoch": 64.0,
"eval_loss": 0.4362615644931793,
"eval_runtime": 10.4425,
"eval_samples_per_second": 478.814,
"eval_steps_per_second": 3.831,
"step": 100000
},
{
"epoch": 64.01,
"learning_rate": 5.330452921628497e-05,
"loss": 0.7108,
"step": 101000
},
{
"epoch": 65.0,
"learning_rate": 5.247918773366112e-05,
"loss": 0.6889,
"step": 102000
},
{
"epoch": 66.0,
"learning_rate": 5.165316846586541e-05,
"loss": 0.7697,
"step": 103000
},
{
"epoch": 66.01,
"learning_rate": 5.0826697238317935e-05,
"loss": 0.6974,
"step": 104000
},
{
"epoch": 67.0,
"learning_rate": 5e-05,
"loss": 0.6968,
"step": 105000
},
{
"epoch": 67.01,
"learning_rate": 4.917330276168208e-05,
"loss": 0.7655,
"step": 106000
},
{
"epoch": 68.0,
"learning_rate": 4.834683153413459e-05,
"loss": 0.6812,
"step": 107000
},
{
"epoch": 69.0,
"learning_rate": 4.7520812266338885e-05,
"loss": 0.7167,
"step": 108000
},
{
"epoch": 69.01,
"learning_rate": 4.669547078371504e-05,
"loss": 0.7414,
"step": 109000
},
{
"epoch": 70.0,
"learning_rate": 4.5871032726383386e-05,
"loss": 0.6714,
"step": 110000
},
{
"epoch": 71.0,
"learning_rate": 4.504772348747687e-05,
"loss": 0.7282,
"step": 111000
},
{
"epoch": 71.01,
"learning_rate": 4.4225768151520694e-05,
"loss": 0.7159,
"step": 112000
},
{
"epoch": 72.0,
"learning_rate": 4.3405391432896555e-05,
"loss": 0.6722,
"step": 113000
},
{
"epoch": 73.0,
"learning_rate": 4.2586817614407895e-05,
"loss": 0.7494,
"step": 114000
},
{
"epoch": 73.01,
"learning_rate": 4.17702704859633e-05,
"loss": 0.69,
"step": 115000
},
{
"epoch": 74.0,
"learning_rate": 4.095597328339452e-05,
"loss": 0.6719,
"step": 116000
},
{
"epoch": 75.0,
"learning_rate": 4.0144148627425993e-05,
"loss": 0.7559,
"step": 117000
},
{
"epoch": 75.01,
"learning_rate": 3.933501846281267e-05,
"loss": 0.6806,
"step": 118000
},
{
"epoch": 76.0,
"learning_rate": 3.852880399766243e-05,
"loss": 0.6823,
"step": 119000
},
{
"epoch": 76.01,
"learning_rate": 3.772572564296005e-05,
"loss": 0.747,
"step": 120000
},
{
"epoch": 77.0,
"learning_rate": 3.6926002952309016e-05,
"loss": 0.664,
"step": 121000
},
{
"epoch": 78.0,
"learning_rate": 3.612985456190778e-05,
"loss": 0.704,
"step": 122000
},
{
"epoch": 78.01,
"learning_rate": 3.533749813077677e-05,
"loss": 0.7249,
"step": 123000
},
{
"epoch": 79.0,
"learning_rate": 3.4549150281252636e-05,
"loss": 0.6559,
"step": 124000
},
{
"epoch": 80.0,
"learning_rate": 3.3765026539765834e-05,
"loss": 0.7263,
"step": 125000
},
{
"epoch": 80.01,
"learning_rate": 3.298534127791785e-05,
"loss": 0.6889,
"step": 126000
},
{
"epoch": 81.0,
"learning_rate": 3.221030765387417e-05,
"loss": 0.6603,
"step": 127000
},
{
"epoch": 82.0,
"learning_rate": 3.144013755408895e-05,
"loss": 0.7373,
"step": 128000
},
{
"epoch": 82.01,
"learning_rate": 3.0675041535377405e-05,
"loss": 0.6749,
"step": 129000
},
{
"epoch": 83.0,
"learning_rate": 2.991522876735154e-05,
"loss": 0.6619,
"step": 130000
},
{
"epoch": 84.0,
"learning_rate": 2.916090697523549e-05,
"loss": 0.7391,
"step": 131000
},
{
"epoch": 84.01,
"learning_rate": 2.8412282383075363e-05,
"loss": 0.666,
"step": 132000
},
{
"epoch": 85.0,
"learning_rate": 2.766955965735968e-05,
"loss": 0.6779,
"step": 133000
},
{
"epoch": 85.01,
"learning_rate": 2.693294185106562e-05,
"loss": 0.7261,
"step": 134000
},
{
"epoch": 86.0,
"learning_rate": 2.6202630348146324e-05,
"loss": 0.6498,
"step": 135000
},
{
"epoch": 87.0,
"learning_rate": 2.547882480847461e-05,
"loss": 0.6967,
"step": 136000
},
{
"epoch": 87.01,
"learning_rate": 2.476172311325783e-05,
"loss": 0.7051,
"step": 137000
},
{
"epoch": 88.0,
"learning_rate": 2.405152131093926e-05,
"loss": 0.6461,
"step": 138000
},
{
"epoch": 89.0,
"learning_rate": 2.3348413563600325e-05,
"loss": 0.7186,
"step": 139000
},
{
"epoch": 89.01,
"learning_rate": 2.2652592093878666e-05,
"loss": 0.6739,
"step": 140000
},
{
"epoch": 90.0,
"learning_rate": 2.196424713241637e-05,
"loss": 0.6494,
"step": 141000
},
{
"epoch": 91.0,
"learning_rate": 2.128356686585282e-05,
"loss": 0.7282,
"step": 142000
},
{
"epoch": 91.01,
"learning_rate": 2.061073738537635e-05,
"loss": 0.6612,
"step": 143000
},
{
"epoch": 92.0,
"learning_rate": 1.9945942635848748e-05,
"loss": 0.6567,
"step": 144000
},
{
"epoch": 93.0,
"learning_rate": 1.928936436551661e-05,
"loss": 0.7282,
"step": 145000
},
{
"epoch": 93.01,
"learning_rate": 1.8641182076323148e-05,
"loss": 0.649,
"step": 146000
},
{
"epoch": 94.0,
"learning_rate": 1.800157297483417e-05,
"loss": 0.6794,
"step": 147000
},
{
"epoch": 94.01,
"learning_rate": 1.7370711923791567e-05,
"loss": 0.7052,
"step": 148000
},
{
"epoch": 95.0,
"learning_rate": 1.6748771394307585e-05,
"loss": 0.6402,
"step": 149000
},
{
"epoch": 96.0,
"learning_rate": 1.6135921418712956e-05,
"loss": 0.6916,
"step": 150000
},
{
"epoch": 96.0,
"eval_loss": 0.4080168604850769,
"eval_runtime": 11.1106,
"eval_samples_per_second": 450.021,
"eval_steps_per_second": 3.6,
"step": 150000
}
],
"max_steps": 200000,
"num_train_epochs": 9223372036854775807,
"total_flos": 2.52721408044014e+18,
"trial_name": null,
"trial_params": null
}