multiopus / trainer_state.json
TuhinColumbia's picture
Upload trainer_state.json
fcee3a4
raw
history blame
8.78 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9999880000239999,
"global_step": 33333,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.9249992499925e-05,
"loss": 1.253,
"step": 500
},
{
"epoch": 0.03,
"learning_rate": 4.849998499985e-05,
"loss": 1.2489,
"step": 1000
},
{
"epoch": 0.04,
"learning_rate": 4.7749977499775e-05,
"loss": 1.2349,
"step": 1500
},
{
"epoch": 0.06,
"learning_rate": 4.69999699997e-05,
"loss": 1.2296,
"step": 2000
},
{
"epoch": 0.07,
"learning_rate": 4.6249962499625e-05,
"loss": 1.2276,
"step": 2500
},
{
"epoch": 0.09,
"learning_rate": 4.549995499955e-05,
"loss": 1.2195,
"step": 3000
},
{
"epoch": 0.1,
"learning_rate": 4.4749947499475e-05,
"loss": 1.2111,
"step": 3500
},
{
"epoch": 0.12,
"learning_rate": 4.39999399994e-05,
"loss": 1.2111,
"step": 4000
},
{
"epoch": 0.13,
"learning_rate": 4.3249932499324996e-05,
"loss": 1.21,
"step": 4500
},
{
"epoch": 0.15,
"learning_rate": 4.2499924999249994e-05,
"loss": 1.2027,
"step": 5000
},
{
"epoch": 0.16,
"learning_rate": 4.174991749917499e-05,
"loss": 1.2024,
"step": 5500
},
{
"epoch": 0.18,
"learning_rate": 4.09999099991e-05,
"loss": 1.2017,
"step": 6000
},
{
"epoch": 0.19,
"learning_rate": 4.0249902499024996e-05,
"loss": 1.2009,
"step": 6500
},
{
"epoch": 0.21,
"learning_rate": 3.9499894998949995e-05,
"loss": 1.192,
"step": 7000
},
{
"epoch": 0.22,
"learning_rate": 3.8749887498874993e-05,
"loss": 1.1969,
"step": 7500
},
{
"epoch": 0.24,
"learning_rate": 3.799987999879999e-05,
"loss": 1.1949,
"step": 8000
},
{
"epoch": 0.25,
"learning_rate": 3.724987249872499e-05,
"loss": 1.1974,
"step": 8500
},
{
"epoch": 0.27,
"learning_rate": 3.649986499864999e-05,
"loss": 1.187,
"step": 9000
},
{
"epoch": 0.28,
"learning_rate": 3.574985749857499e-05,
"loss": 1.1831,
"step": 9500
},
{
"epoch": 0.3,
"learning_rate": 3.4999849998499986e-05,
"loss": 1.183,
"step": 10000
},
{
"epoch": 0.31,
"learning_rate": 3.4249842498424984e-05,
"loss": 1.1819,
"step": 10500
},
{
"epoch": 0.33,
"learning_rate": 3.349983499834998e-05,
"loss": 1.1824,
"step": 11000
},
{
"epoch": 0.34,
"learning_rate": 3.274982749827498e-05,
"loss": 1.1832,
"step": 11500
},
{
"epoch": 0.36,
"learning_rate": 3.199981999819998e-05,
"loss": 1.1781,
"step": 12000
},
{
"epoch": 0.37,
"learning_rate": 3.124981249812498e-05,
"loss": 1.1775,
"step": 12500
},
{
"epoch": 0.39,
"learning_rate": 3.049980499804998e-05,
"loss": 1.1699,
"step": 13000
},
{
"epoch": 0.4,
"learning_rate": 2.974979749797498e-05,
"loss": 1.1687,
"step": 13500
},
{
"epoch": 0.42,
"learning_rate": 2.8999789997899977e-05,
"loss": 1.1653,
"step": 14000
},
{
"epoch": 0.43,
"learning_rate": 2.824978249782498e-05,
"loss": 1.1712,
"step": 14500
},
{
"epoch": 0.45,
"learning_rate": 2.7499774997749978e-05,
"loss": 1.1671,
"step": 15000
},
{
"epoch": 0.46,
"learning_rate": 2.6749767497674976e-05,
"loss": 1.1634,
"step": 15500
},
{
"epoch": 0.48,
"learning_rate": 2.5999759997599975e-05,
"loss": 1.1667,
"step": 16000
},
{
"epoch": 0.49,
"learning_rate": 2.5249752497524977e-05,
"loss": 1.159,
"step": 16500
},
{
"epoch": 0.51,
"learning_rate": 2.4499744997449975e-05,
"loss": 1.1569,
"step": 17000
},
{
"epoch": 0.52,
"learning_rate": 2.3749737497374974e-05,
"loss": 1.1556,
"step": 17500
},
{
"epoch": 0.54,
"learning_rate": 2.2999729997299972e-05,
"loss": 1.1603,
"step": 18000
},
{
"epoch": 0.55,
"learning_rate": 2.2249722497224974e-05,
"loss": 1.1533,
"step": 18500
},
{
"epoch": 0.57,
"learning_rate": 2.1499714997149973e-05,
"loss": 1.1536,
"step": 19000
},
{
"epoch": 0.58,
"learning_rate": 2.074970749707497e-05,
"loss": 1.1563,
"step": 19500
},
{
"epoch": 0.6,
"learning_rate": 1.999969999699997e-05,
"loss": 1.1494,
"step": 20000
},
{
"epoch": 0.61,
"learning_rate": 1.924969249692497e-05,
"loss": 1.1511,
"step": 20500
},
{
"epoch": 0.63,
"learning_rate": 1.849968499684997e-05,
"loss": 1.1432,
"step": 21000
},
{
"epoch": 0.64,
"learning_rate": 1.774967749677497e-05,
"loss": 1.1525,
"step": 21500
},
{
"epoch": 0.66,
"learning_rate": 1.6999669996699967e-05,
"loss": 1.1437,
"step": 22000
},
{
"epoch": 0.67,
"learning_rate": 1.624966249662497e-05,
"loss": 1.1452,
"step": 22500
},
{
"epoch": 0.69,
"learning_rate": 1.5499654996549967e-05,
"loss": 1.1411,
"step": 23000
},
{
"epoch": 0.7,
"learning_rate": 1.4749647496474966e-05,
"loss": 1.1381,
"step": 23500
},
{
"epoch": 0.72,
"learning_rate": 1.3999639996399966e-05,
"loss": 1.1412,
"step": 24000
},
{
"epoch": 0.73,
"learning_rate": 1.3249632496324964e-05,
"loss": 1.1405,
"step": 24500
},
{
"epoch": 0.75,
"learning_rate": 1.2499624996249963e-05,
"loss": 1.1418,
"step": 25000
},
{
"epoch": 0.76,
"learning_rate": 1.1749617496174963e-05,
"loss": 1.1351,
"step": 25500
},
{
"epoch": 0.78,
"learning_rate": 1.0999609996099962e-05,
"loss": 1.1364,
"step": 26000
},
{
"epoch": 0.79,
"learning_rate": 1.0249602496024962e-05,
"loss": 1.1322,
"step": 26500
},
{
"epoch": 0.81,
"learning_rate": 9.49959499594996e-06,
"loss": 1.1337,
"step": 27000
},
{
"epoch": 0.82,
"learning_rate": 8.749587495874959e-06,
"loss": 1.1322,
"step": 27500
},
{
"epoch": 0.84,
"learning_rate": 7.999579995799957e-06,
"loss": 1.128,
"step": 28000
},
{
"epoch": 0.85,
"learning_rate": 7.249572495724958e-06,
"loss": 1.1318,
"step": 28500
},
{
"epoch": 0.87,
"learning_rate": 6.499564995649957e-06,
"loss": 1.1283,
"step": 29000
},
{
"epoch": 0.88,
"learning_rate": 5.749557495574956e-06,
"loss": 1.1262,
"step": 29500
},
{
"epoch": 0.9,
"learning_rate": 4.999549995499956e-06,
"loss": 1.1295,
"step": 30000
},
{
"epoch": 0.91,
"learning_rate": 4.249542495424955e-06,
"loss": 1.125,
"step": 30500
},
{
"epoch": 0.93,
"learning_rate": 3.499534995349954e-06,
"loss": 1.1241,
"step": 31000
},
{
"epoch": 0.94,
"learning_rate": 2.749527495274953e-06,
"loss": 1.1155,
"step": 31500
},
{
"epoch": 0.96,
"learning_rate": 1.999519995199952e-06,
"loss": 1.1226,
"step": 32000
},
{
"epoch": 0.97,
"learning_rate": 1.2495124951249513e-06,
"loss": 1.1281,
"step": 32500
},
{
"epoch": 0.99,
"learning_rate": 4.995049950499505e-07,
"loss": 1.1218,
"step": 33000
},
{
"epoch": 1.0,
"eval_bleu": 33.5469,
"eval_gen_len": 17.7505,
"eval_loss": 1.3453229665756226,
"eval_runtime": 449.9191,
"eval_samples_per_second": 4.445,
"step": 33333
},
{
"epoch": 1.0,
"step": 33333,
"total_flos": 2.7716100970398106e+18,
"train_runtime": 119333.5912,
"train_samples_per_second": 0.279
}
],
"max_steps": 33333,
"num_train_epochs": 1,
"total_flos": 2.7716100970398106e+18,
"trial_name": null,
"trial_params": null
}