roberta-clinical-wl-es / trainer_state.json
caracena
initial commit
14ef454
raw
history blame
12.3 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"global_step": 48264,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 4.9740999740999744e-05,
"loss": 1.3692,
"step": 500
},
{
"epoch": 0.03,
"learning_rate": 4.9481999481999486e-05,
"loss": 1.1369,
"step": 1000
},
{
"epoch": 0.05,
"learning_rate": 4.922299922299923e-05,
"loss": 1.0305,
"step": 1500
},
{
"epoch": 0.06,
"learning_rate": 4.896399896399896e-05,
"loss": 0.9746,
"step": 2000
},
{
"epoch": 0.08,
"learning_rate": 4.8704998704998705e-05,
"loss": 0.9232,
"step": 2500
},
{
"epoch": 0.09,
"learning_rate": 4.844599844599845e-05,
"loss": 0.9313,
"step": 3000
},
{
"epoch": 0.11,
"learning_rate": 4.818699818699819e-05,
"loss": 0.8694,
"step": 3500
},
{
"epoch": 0.12,
"learning_rate": 4.792799792799793e-05,
"loss": 0.8683,
"step": 4000
},
{
"epoch": 0.14,
"learning_rate": 4.766899766899767e-05,
"loss": 0.8391,
"step": 4500
},
{
"epoch": 0.31,
"learning_rate": 4.482015580971324e-05,
"loss": 0.8134,
"step": 5000
},
{
"epoch": 0.34,
"learning_rate": 4.4302171390684573e-05,
"loss": 0.7795,
"step": 5500
},
{
"epoch": 0.37,
"learning_rate": 4.37841869716559e-05,
"loss": 0.7527,
"step": 6000
},
{
"epoch": 0.4,
"learning_rate": 4.3266202552627213e-05,
"loss": 0.7455,
"step": 6500
},
{
"epoch": 0.44,
"learning_rate": 4.2748218133598544e-05,
"loss": 0.7321,
"step": 7000
},
{
"epoch": 0.47,
"learning_rate": 4.223023371456987e-05,
"loss": 0.7099,
"step": 7500
},
{
"epoch": 0.5,
"learning_rate": 4.17122492955412e-05,
"loss": 0.7011,
"step": 8000
},
{
"epoch": 0.53,
"learning_rate": 4.1194264876512514e-05,
"loss": 0.685,
"step": 8500
},
{
"epoch": 0.56,
"learning_rate": 4.067628045748384e-05,
"loss": 0.6889,
"step": 9000
},
{
"epoch": 0.59,
"learning_rate": 4.015829603845517e-05,
"loss": 0.6752,
"step": 9500
},
{
"epoch": 0.62,
"learning_rate": 3.964031161942649e-05,
"loss": 0.6598,
"step": 10000
},
{
"epoch": 0.65,
"learning_rate": 3.9122327200397814e-05,
"loss": 0.6626,
"step": 10500
},
{
"epoch": 0.68,
"learning_rate": 3.860434278136914e-05,
"loss": 0.6569,
"step": 11000
},
{
"epoch": 0.71,
"learning_rate": 3.808635836234046e-05,
"loss": 0.6452,
"step": 11500
},
{
"epoch": 0.75,
"learning_rate": 3.7568373943311785e-05,
"loss": 0.6412,
"step": 12000
},
{
"epoch": 0.78,
"learning_rate": 3.7050389524283115e-05,
"loss": 0.6349,
"step": 12500
},
{
"epoch": 0.81,
"learning_rate": 3.653240510525444e-05,
"loss": 0.6315,
"step": 13000
},
{
"epoch": 0.84,
"learning_rate": 3.601442068622576e-05,
"loss": 0.636,
"step": 13500
},
{
"epoch": 0.87,
"learning_rate": 3.5496436267197085e-05,
"loss": 0.621,
"step": 14000
},
{
"epoch": 0.9,
"learning_rate": 3.497845184816841e-05,
"loss": 0.6226,
"step": 14500
},
{
"epoch": 0.93,
"learning_rate": 3.446046742913973e-05,
"loss": 0.6188,
"step": 15000
},
{
"epoch": 0.96,
"learning_rate": 3.394248301011106e-05,
"loss": 0.6055,
"step": 15500
},
{
"epoch": 0.99,
"learning_rate": 3.342449859108238e-05,
"loss": 0.6105,
"step": 16000
},
{
"epoch": 1.03,
"learning_rate": 3.29065141720537e-05,
"loss": 0.5962,
"step": 16500
},
{
"epoch": 1.06,
"learning_rate": 3.238852975302503e-05,
"loss": 0.5865,
"step": 17000
},
{
"epoch": 1.09,
"learning_rate": 3.1870545333996356e-05,
"loss": 0.5869,
"step": 17500
},
{
"epoch": 1.12,
"learning_rate": 3.135256091496768e-05,
"loss": 0.5868,
"step": 18000
},
{
"epoch": 1.15,
"learning_rate": 3.0834576495939e-05,
"loss": 0.5924,
"step": 18500
},
{
"epoch": 1.18,
"learning_rate": 3.0316592076910326e-05,
"loss": 0.5798,
"step": 19000
},
{
"epoch": 1.21,
"learning_rate": 2.979860765788165e-05,
"loss": 0.574,
"step": 19500
},
{
"epoch": 1.24,
"learning_rate": 2.928062323885298e-05,
"loss": 0.5684,
"step": 20000
},
{
"epoch": 1.27,
"learning_rate": 2.87626388198243e-05,
"loss": 0.5529,
"step": 20500
},
{
"epoch": 1.31,
"learning_rate": 2.824465440079563e-05,
"loss": 0.5733,
"step": 21000
},
{
"epoch": 1.34,
"learning_rate": 2.772666998176695e-05,
"loss": 0.5594,
"step": 21500
},
{
"epoch": 1.37,
"learning_rate": 2.7208685562738273e-05,
"loss": 0.5557,
"step": 22000
},
{
"epoch": 1.4,
"learning_rate": 2.66907011437096e-05,
"loss": 0.5595,
"step": 22500
},
{
"epoch": 1.43,
"learning_rate": 2.6172716724680923e-05,
"loss": 0.5519,
"step": 23000
},
{
"epoch": 1.46,
"learning_rate": 2.5654732305652247e-05,
"loss": 0.5569,
"step": 23500
},
{
"epoch": 1.49,
"learning_rate": 2.5136747886623573e-05,
"loss": 0.5457,
"step": 24000
},
{
"epoch": 1.52,
"learning_rate": 2.4618763467594897e-05,
"loss": 0.5511,
"step": 24500
},
{
"epoch": 1.55,
"learning_rate": 2.410077904856622e-05,
"loss": 0.5547,
"step": 25000
},
{
"epoch": 1.59,
"learning_rate": 2.3582794629537544e-05,
"loss": 0.5384,
"step": 25500
},
{
"epoch": 1.62,
"learning_rate": 2.306481021050887e-05,
"loss": 0.5353,
"step": 26000
},
{
"epoch": 1.65,
"learning_rate": 2.2546825791480194e-05,
"loss": 0.5226,
"step": 26500
},
{
"epoch": 1.68,
"learning_rate": 2.2028841372451517e-05,
"loss": 0.5346,
"step": 27000
},
{
"epoch": 1.71,
"learning_rate": 2.151085695342284e-05,
"loss": 0.5323,
"step": 27500
},
{
"epoch": 1.74,
"learning_rate": 2.0992872534394168e-05,
"loss": 0.5399,
"step": 28000
},
{
"epoch": 1.77,
"learning_rate": 2.047488811536549e-05,
"loss": 0.5269,
"step": 28500
},
{
"epoch": 1.8,
"learning_rate": 1.9956903696336814e-05,
"loss": 0.5348,
"step": 29000
},
{
"epoch": 1.83,
"learning_rate": 1.943891927730814e-05,
"loss": 0.5273,
"step": 29500
},
{
"epoch": 1.86,
"learning_rate": 1.892093485827946e-05,
"loss": 0.5239,
"step": 30000
},
{
"epoch": 1.9,
"learning_rate": 1.8402950439250788e-05,
"loss": 0.5219,
"step": 30500
},
{
"epoch": 1.93,
"learning_rate": 1.7884966020222115e-05,
"loss": 0.5224,
"step": 31000
},
{
"epoch": 1.96,
"learning_rate": 1.7366981601193438e-05,
"loss": 0.5255,
"step": 31500
},
{
"epoch": 1.99,
"learning_rate": 1.684899718216476e-05,
"loss": 0.5115,
"step": 32000
},
{
"epoch": 2.02,
"learning_rate": 1.6331012763136085e-05,
"loss": 0.5238,
"step": 32500
},
{
"epoch": 2.05,
"learning_rate": 1.5813028344107412e-05,
"loss": 0.5124,
"step": 33000
},
{
"epoch": 2.08,
"learning_rate": 1.5295043925078735e-05,
"loss": 0.5217,
"step": 33500
},
{
"epoch": 2.11,
"learning_rate": 1.4777059506050059e-05,
"loss": 0.5122,
"step": 34000
},
{
"epoch": 2.14,
"learning_rate": 1.4259075087021384e-05,
"loss": 0.504,
"step": 34500
},
{
"epoch": 2.18,
"learning_rate": 1.3741090667992707e-05,
"loss": 0.5041,
"step": 35000
},
{
"epoch": 2.21,
"learning_rate": 1.3223106248964032e-05,
"loss": 0.5082,
"step": 35500
},
{
"epoch": 2.24,
"learning_rate": 1.2705121829935357e-05,
"loss": 0.5005,
"step": 36000
},
{
"epoch": 2.27,
"learning_rate": 1.218713741090668e-05,
"loss": 0.4966,
"step": 36500
},
{
"epoch": 2.3,
"learning_rate": 1.1669152991878004e-05,
"loss": 0.5147,
"step": 37000
},
{
"epoch": 2.33,
"learning_rate": 1.115116857284933e-05,
"loss": 0.5101,
"step": 37500
},
{
"epoch": 2.36,
"learning_rate": 1.0633184153820654e-05,
"loss": 0.501,
"step": 38000
},
{
"epoch": 2.39,
"learning_rate": 1.0115199734791978e-05,
"loss": 0.5042,
"step": 38500
},
{
"epoch": 2.42,
"learning_rate": 9.597215315763303e-06,
"loss": 0.4999,
"step": 39000
},
{
"epoch": 2.46,
"learning_rate": 9.079230896734626e-06,
"loss": 0.4954,
"step": 39500
},
{
"epoch": 2.49,
"learning_rate": 8.56124647770595e-06,
"loss": 0.4924,
"step": 40000
},
{
"epoch": 2.52,
"learning_rate": 8.043262058677275e-06,
"loss": 0.4882,
"step": 40500
},
{
"epoch": 2.55,
"learning_rate": 7.525277639648599e-06,
"loss": 0.495,
"step": 41000
},
{
"epoch": 2.58,
"learning_rate": 7.007293220619925e-06,
"loss": 0.5013,
"step": 41500
},
{
"epoch": 2.61,
"learning_rate": 6.489308801591248e-06,
"loss": 0.4853,
"step": 42000
},
{
"epoch": 2.64,
"learning_rate": 5.971324382562573e-06,
"loss": 0.5026,
"step": 42500
},
{
"epoch": 2.67,
"learning_rate": 5.453339963533898e-06,
"loss": 0.4927,
"step": 43000
},
{
"epoch": 2.7,
"learning_rate": 4.935355544505221e-06,
"loss": 0.49,
"step": 43500
},
{
"epoch": 2.73,
"learning_rate": 4.417371125476545e-06,
"loss": 0.483,
"step": 44000
},
{
"epoch": 2.77,
"learning_rate": 3.8993867064478705e-06,
"loss": 0.4942,
"step": 44500
},
{
"epoch": 2.8,
"learning_rate": 3.3814022874191948e-06,
"loss": 0.4902,
"step": 45000
},
{
"epoch": 2.83,
"learning_rate": 2.863417868390519e-06,
"loss": 0.486,
"step": 45500
},
{
"epoch": 2.86,
"learning_rate": 2.3454334493618433e-06,
"loss": 0.4862,
"step": 46000
},
{
"epoch": 2.89,
"learning_rate": 1.8274490303331678e-06,
"loss": 0.485,
"step": 46500
},
{
"epoch": 2.92,
"learning_rate": 1.309464611304492e-06,
"loss": 0.4851,
"step": 47000
},
{
"epoch": 2.95,
"learning_rate": 7.914801922758165e-07,
"loss": 0.4909,
"step": 47500
},
{
"epoch": 2.98,
"learning_rate": 2.7349577324714074e-07,
"loss": 0.4866,
"step": 48000
},
{
"epoch": 3.0,
"step": 48264,
"total_flos": 1.9381442927896166e+17,
"train_loss": 0.511344972582336,
"train_runtime": 8035.7257,
"train_samples_per_second": 96.093,
"train_steps_per_second": 6.006
}
],
"max_steps": 48264,
"num_train_epochs": 3,
"total_flos": 1.9381442927896166e+17,
"trial_name": null,
"trial_params": null
}