longbartwithdoctype / trainer_state.json
BeumJin's picture
longformer doctype
95eeba5
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 6.999985466172516,
"global_step": 120407,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.12,
"learning_rate": 2e-05,
"loss": 9.0919,
"step": 2000
},
{
"epoch": 0.23,
"learning_rate": 4e-05,
"loss": 7.6432,
"step": 4000
},
{
"epoch": 0.35,
"learning_rate": 6e-05,
"loss": 6.7199,
"step": 6000
},
{
"epoch": 0.47,
"learning_rate": 8e-05,
"loss": 6.1549,
"step": 8000
},
{
"epoch": 0.58,
"learning_rate": 0.0001,
"loss": 5.5391,
"step": 10000
},
{
"epoch": 0.7,
"learning_rate": 9.818852065539323e-05,
"loss": 4.4909,
"step": 12000
},
{
"epoch": 0.81,
"learning_rate": 9.637704131078646e-05,
"loss": 2.7203,
"step": 14000
},
{
"epoch": 0.93,
"learning_rate": 9.456556196617969e-05,
"loss": 1.5493,
"step": 16000
},
{
"epoch": 1.05,
"learning_rate": 9.275408262157291e-05,
"loss": 1.2876,
"step": 18000
},
{
"epoch": 1.16,
"learning_rate": 9.094260327696614e-05,
"loss": 1.1713,
"step": 20000
},
{
"epoch": 1.28,
"learning_rate": 8.913112393235937e-05,
"loss": 1.1049,
"step": 22000
},
{
"epoch": 1.4,
"learning_rate": 8.73196445877526e-05,
"loss": 1.0576,
"step": 24000
},
{
"epoch": 1.51,
"learning_rate": 8.550816524314582e-05,
"loss": 1.0238,
"step": 26000
},
{
"epoch": 1.63,
"learning_rate": 8.369668589853905e-05,
"loss": 0.9882,
"step": 28000
},
{
"epoch": 1.74,
"learning_rate": 8.188520655393228e-05,
"loss": 0.9338,
"step": 30000
},
{
"epoch": 1.86,
"learning_rate": 8.00737272093255e-05,
"loss": 0.895,
"step": 32000
},
{
"epoch": 1.98,
"learning_rate": 7.826224786471873e-05,
"loss": 0.8748,
"step": 34000
},
{
"epoch": 2.09,
"learning_rate": 7.645076852011194e-05,
"loss": 0.8617,
"step": 36000
},
{
"epoch": 2.21,
"learning_rate": 7.463928917550518e-05,
"loss": 0.8464,
"step": 38000
},
{
"epoch": 2.33,
"learning_rate": 7.282780983089841e-05,
"loss": 0.8373,
"step": 40000
},
{
"epoch": 2.44,
"learning_rate": 7.101633048629164e-05,
"loss": 0.8292,
"step": 42000
},
{
"epoch": 2.56,
"learning_rate": 6.920485114168485e-05,
"loss": 0.8215,
"step": 44000
},
{
"epoch": 2.67,
"learning_rate": 6.739337179707809e-05,
"loss": 0.8139,
"step": 46000
},
{
"epoch": 2.79,
"learning_rate": 6.558189245247132e-05,
"loss": 0.8065,
"step": 48000
},
{
"epoch": 2.91,
"learning_rate": 6.377041310786455e-05,
"loss": 0.8003,
"step": 50000
},
{
"epoch": 3.02,
"learning_rate": 6.195893376325776e-05,
"loss": 0.7953,
"step": 52000
},
{
"epoch": 3.14,
"learning_rate": 6.0147454418650995e-05,
"loss": 0.7864,
"step": 54000
},
{
"epoch": 3.26,
"learning_rate": 5.833597507404422e-05,
"loss": 0.7826,
"step": 56000
},
{
"epoch": 3.37,
"learning_rate": 5.652449572943744e-05,
"loss": 0.7781,
"step": 58000
},
{
"epoch": 3.49,
"learning_rate": 5.4713016384830676e-05,
"loss": 0.7744,
"step": 60000
},
{
"epoch": 3.6,
"learning_rate": 5.2901537040223903e-05,
"loss": 0.7678,
"step": 62000
},
{
"epoch": 3.72,
"learning_rate": 5.109005769561713e-05,
"loss": 0.7646,
"step": 64000
},
{
"epoch": 3.84,
"learning_rate": 4.927857835101036e-05,
"loss": 0.7621,
"step": 66000
},
{
"epoch": 3.95,
"learning_rate": 4.746709900640358e-05,
"loss": 0.7588,
"step": 68000
},
{
"epoch": 4.07,
"learning_rate": 4.565561966179681e-05,
"loss": 0.7535,
"step": 70000
},
{
"epoch": 4.19,
"learning_rate": 4.384414031719003e-05,
"loss": 0.7494,
"step": 72000
},
{
"epoch": 4.3,
"learning_rate": 4.203266097258326e-05,
"loss": 0.7466,
"step": 74000
},
{
"epoch": 4.42,
"learning_rate": 4.022118162797649e-05,
"loss": 0.7426,
"step": 76000
},
{
"epoch": 4.53,
"learning_rate": 3.8409702283369714e-05,
"loss": 0.7402,
"step": 78000
},
{
"epoch": 4.65,
"learning_rate": 3.659822293876294e-05,
"loss": 0.7386,
"step": 80000
},
{
"epoch": 4.77,
"learning_rate": 3.478674359415617e-05,
"loss": 0.7348,
"step": 82000
},
{
"epoch": 4.88,
"learning_rate": 3.2975264249549395e-05,
"loss": 0.7331,
"step": 84000
},
{
"epoch": 5.0,
"learning_rate": 3.116378490494262e-05,
"loss": 0.7307,
"step": 86000
},
{
"epoch": 5.12,
"learning_rate": 2.9352305560335853e-05,
"loss": 0.7258,
"step": 88000
},
{
"epoch": 5.23,
"learning_rate": 2.7540826215729077e-05,
"loss": 0.7234,
"step": 90000
},
{
"epoch": 5.35,
"learning_rate": 2.57293468711223e-05,
"loss": 0.7208,
"step": 92000
},
{
"epoch": 5.46,
"learning_rate": 2.391786752651553e-05,
"loss": 0.7179,
"step": 94000
},
{
"epoch": 5.58,
"learning_rate": 2.2106388181908758e-05,
"loss": 0.7188,
"step": 96000
},
{
"epoch": 5.7,
"learning_rate": 2.0294908837301985e-05,
"loss": 0.7158,
"step": 98000
},
{
"epoch": 5.81,
"learning_rate": 1.8483429492695212e-05,
"loss": 0.7146,
"step": 100000
},
{
"epoch": 5.93,
"learning_rate": 1.6671950148088436e-05,
"loss": 0.7125,
"step": 102000
},
{
"epoch": 6.05,
"learning_rate": 1.4860470803481663e-05,
"loss": 0.7092,
"step": 104000
},
{
"epoch": 6.16,
"learning_rate": 1.304899145887489e-05,
"loss": 0.7061,
"step": 106000
},
{
"epoch": 6.28,
"learning_rate": 1.1237512114268117e-05,
"loss": 0.7064,
"step": 108000
},
{
"epoch": 6.39,
"learning_rate": 9.426032769661344e-06,
"loss": 0.7034,
"step": 110000
},
{
"epoch": 6.51,
"learning_rate": 7.614553425054572e-06,
"loss": 0.7027,
"step": 112000
},
{
"epoch": 6.63,
"learning_rate": 5.803074080447798e-06,
"loss": 0.7015,
"step": 114000
},
{
"epoch": 6.74,
"learning_rate": 3.991594735841025e-06,
"loss": 0.7004,
"step": 116000
},
{
"epoch": 6.86,
"learning_rate": 2.1801153912342513e-06,
"loss": 0.7,
"step": 118000
},
{
"epoch": 6.98,
"learning_rate": 3.6863604662747836e-07,
"loss": 0.6989,
"step": 120000
},
{
"epoch": 7.0,
"step": 120407,
"total_flos": 3.3047861122616525e+17,
"train_loss": 1.4238887095158894,
"train_runtime": 60363.921,
"train_samples_per_second": 31.915,
"train_steps_per_second": 1.995
}
],
"max_steps": 120407,
"num_train_epochs": 7,
"total_flos": 3.3047861122616525e+17,
"trial_name": null,
"trial_params": null
}