mbert-resp-en-it / trainer_state.json
Nikita Moghe Moghe
First model version
87da19a
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 20.0,
"global_step": 250000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.2,
"learning_rate": 4.9500000000000004e-05,
"loss": 5.8035,
"step": 2500
},
{
"epoch": 0.4,
"learning_rate": 4.9e-05,
"loss": 5.6846,
"step": 5000
},
{
"epoch": 0.6,
"learning_rate": 4.85e-05,
"loss": 5.6715,
"step": 7500
},
{
"epoch": 0.8,
"learning_rate": 4.8e-05,
"loss": 5.6443,
"step": 10000
},
{
"epoch": 1.0,
"learning_rate": 4.75e-05,
"loss": 5.6103,
"step": 12500
},
{
"epoch": 1.2,
"learning_rate": 4.7e-05,
"loss": 5.5002,
"step": 15000
},
{
"epoch": 1.4,
"learning_rate": 4.6500000000000005e-05,
"loss": 5.4878,
"step": 17500
},
{
"epoch": 1.6,
"learning_rate": 4.600000000000001e-05,
"loss": 5.5005,
"step": 20000
},
{
"epoch": 1.8,
"learning_rate": 4.55e-05,
"loss": 5.5005,
"step": 22500
},
{
"epoch": 2.0,
"learning_rate": 4.5e-05,
"loss": 5.4821,
"step": 25000
},
{
"epoch": 2.2,
"learning_rate": 4.4500000000000004e-05,
"loss": 5.3239,
"step": 27500
},
{
"epoch": 2.4,
"learning_rate": 4.4000000000000006e-05,
"loss": 5.3328,
"step": 30000
},
{
"epoch": 2.6,
"learning_rate": 4.35e-05,
"loss": 5.3409,
"step": 32500
},
{
"epoch": 2.8,
"learning_rate": 4.3e-05,
"loss": 5.3499,
"step": 35000
},
{
"epoch": 3.0,
"learning_rate": 4.25e-05,
"loss": 5.3475,
"step": 37500
},
{
"epoch": 3.2,
"learning_rate": 4.2e-05,
"loss": 5.1148,
"step": 40000
},
{
"epoch": 3.4,
"learning_rate": 4.15e-05,
"loss": 5.122,
"step": 42500
},
{
"epoch": 3.6,
"learning_rate": 4.1e-05,
"loss": 5.1339,
"step": 45000
},
{
"epoch": 3.8,
"learning_rate": 4.05e-05,
"loss": 5.1598,
"step": 47500
},
{
"epoch": 4.0,
"learning_rate": 4e-05,
"loss": 5.1829,
"step": 50000
},
{
"epoch": 4.2,
"learning_rate": 3.9500000000000005e-05,
"loss": 4.8405,
"step": 52500
},
{
"epoch": 4.4,
"learning_rate": 3.9000000000000006e-05,
"loss": 4.8717,
"step": 55000
},
{
"epoch": 4.6,
"learning_rate": 3.85e-05,
"loss": 4.9215,
"step": 57500
},
{
"epoch": 4.8,
"learning_rate": 3.8e-05,
"loss": 4.9175,
"step": 60000
},
{
"epoch": 5.0,
"learning_rate": 3.7500000000000003e-05,
"loss": 4.9463,
"step": 62500
},
{
"epoch": 5.2,
"learning_rate": 3.7e-05,
"loss": 4.5458,
"step": 65000
},
{
"epoch": 5.4,
"learning_rate": 3.65e-05,
"loss": 4.5754,
"step": 67500
},
{
"epoch": 5.6,
"learning_rate": 3.6e-05,
"loss": 4.6175,
"step": 70000
},
{
"epoch": 5.8,
"learning_rate": 3.55e-05,
"loss": 4.6592,
"step": 72500
},
{
"epoch": 6.0,
"learning_rate": 3.5e-05,
"loss": 4.6839,
"step": 75000
},
{
"epoch": 6.2,
"learning_rate": 3.45e-05,
"loss": 4.2337,
"step": 77500
},
{
"epoch": 6.4,
"learning_rate": 3.4000000000000007e-05,
"loss": 4.2718,
"step": 80000
},
{
"epoch": 6.6,
"learning_rate": 3.35e-05,
"loss": 4.3148,
"step": 82500
},
{
"epoch": 6.8,
"learning_rate": 3.3e-05,
"loss": 4.3662,
"step": 85000
},
{
"epoch": 7.0,
"learning_rate": 3.2500000000000004e-05,
"loss": 4.3476,
"step": 87500
},
{
"epoch": 7.2,
"learning_rate": 3.2000000000000005e-05,
"loss": 3.894,
"step": 90000
},
{
"epoch": 7.4,
"learning_rate": 3.15e-05,
"loss": 3.9493,
"step": 92500
},
{
"epoch": 7.6,
"learning_rate": 3.1e-05,
"loss": 4.0003,
"step": 95000
},
{
"epoch": 7.8,
"learning_rate": 3.05e-05,
"loss": 4.0301,
"step": 97500
},
{
"epoch": 8.0,
"learning_rate": 3e-05,
"loss": 4.0559,
"step": 100000
},
{
"epoch": 8.2,
"learning_rate": 2.95e-05,
"loss": 3.5674,
"step": 102500
},
{
"epoch": 8.4,
"learning_rate": 2.9e-05,
"loss": 3.6282,
"step": 105000
},
{
"epoch": 8.6,
"learning_rate": 2.8499999999999998e-05,
"loss": 3.6712,
"step": 107500
},
{
"epoch": 8.8,
"learning_rate": 2.8000000000000003e-05,
"loss": 3.7138,
"step": 110000
},
{
"epoch": 9.0,
"learning_rate": 2.7500000000000004e-05,
"loss": 3.7451,
"step": 112500
},
{
"epoch": 9.2,
"learning_rate": 2.7000000000000002e-05,
"loss": 3.2761,
"step": 115000
},
{
"epoch": 9.4,
"learning_rate": 2.6500000000000004e-05,
"loss": 3.3254,
"step": 117500
},
{
"epoch": 9.6,
"learning_rate": 2.6000000000000002e-05,
"loss": 3.3704,
"step": 120000
},
{
"epoch": 9.8,
"learning_rate": 2.5500000000000003e-05,
"loss": 3.375,
"step": 122500
},
{
"epoch": 10.0,
"learning_rate": 2.5e-05,
"loss": 3.4389,
"step": 125000
},
{
"epoch": 10.2,
"learning_rate": 2.45e-05,
"loss": 2.9809,
"step": 127500
},
{
"epoch": 10.4,
"learning_rate": 2.4e-05,
"loss": 3.0355,
"step": 130000
},
{
"epoch": 10.6,
"learning_rate": 2.35e-05,
"loss": 3.103,
"step": 132500
},
{
"epoch": 10.8,
"learning_rate": 2.3000000000000003e-05,
"loss": 3.0881,
"step": 135000
},
{
"epoch": 11.0,
"learning_rate": 2.25e-05,
"loss": 3.144,
"step": 137500
},
{
"epoch": 11.2,
"learning_rate": 2.2000000000000003e-05,
"loss": 2.7198,
"step": 140000
},
{
"epoch": 11.4,
"learning_rate": 2.15e-05,
"loss": 2.7712,
"step": 142500
},
{
"epoch": 11.6,
"learning_rate": 2.1e-05,
"loss": 2.8035,
"step": 145000
},
{
"epoch": 11.8,
"learning_rate": 2.05e-05,
"loss": 2.8412,
"step": 147500
},
{
"epoch": 12.0,
"learning_rate": 2e-05,
"loss": 2.8714,
"step": 150000
},
{
"epoch": 12.2,
"learning_rate": 1.9500000000000003e-05,
"loss": 2.4583,
"step": 152500
},
{
"epoch": 12.4,
"learning_rate": 1.9e-05,
"loss": 2.5254,
"step": 155000
},
{
"epoch": 12.6,
"learning_rate": 1.85e-05,
"loss": 2.5436,
"step": 157500
},
{
"epoch": 12.8,
"learning_rate": 1.8e-05,
"loss": 2.6066,
"step": 160000
},
{
"epoch": 13.0,
"learning_rate": 1.75e-05,
"loss": 2.6249,
"step": 162500
},
{
"epoch": 13.2,
"learning_rate": 1.7000000000000003e-05,
"loss": 2.2585,
"step": 165000
},
{
"epoch": 13.4,
"learning_rate": 1.65e-05,
"loss": 2.3274,
"step": 167500
},
{
"epoch": 13.6,
"learning_rate": 1.6000000000000003e-05,
"loss": 2.3387,
"step": 170000
},
{
"epoch": 13.8,
"learning_rate": 1.55e-05,
"loss": 2.3526,
"step": 172500
},
{
"epoch": 14.0,
"learning_rate": 1.5e-05,
"loss": 2.3839,
"step": 175000
},
{
"epoch": 14.2,
"learning_rate": 1.45e-05,
"loss": 2.0807,
"step": 177500
},
{
"epoch": 14.4,
"learning_rate": 1.4000000000000001e-05,
"loss": 2.1081,
"step": 180000
},
{
"epoch": 14.6,
"learning_rate": 1.3500000000000001e-05,
"loss": 2.1325,
"step": 182500
},
{
"epoch": 14.8,
"learning_rate": 1.3000000000000001e-05,
"loss": 2.1671,
"step": 185000
},
{
"epoch": 15.0,
"learning_rate": 1.25e-05,
"loss": 2.1625,
"step": 187500
},
{
"epoch": 15.2,
"learning_rate": 1.2e-05,
"loss": 1.9216,
"step": 190000
},
{
"epoch": 15.4,
"learning_rate": 1.1500000000000002e-05,
"loss": 1.9387,
"step": 192500
},
{
"epoch": 15.6,
"learning_rate": 1.1000000000000001e-05,
"loss": 1.9563,
"step": 195000
},
{
"epoch": 15.8,
"learning_rate": 1.05e-05,
"loss": 1.9744,
"step": 197500
},
{
"epoch": 16.0,
"learning_rate": 1e-05,
"loss": 1.9931,
"step": 200000
},
{
"epoch": 16.2,
"learning_rate": 9.5e-06,
"loss": 1.7733,
"step": 202500
},
{
"epoch": 16.4,
"learning_rate": 9e-06,
"loss": 1.7925,
"step": 205000
},
{
"epoch": 16.6,
"learning_rate": 8.500000000000002e-06,
"loss": 1.7901,
"step": 207500
},
{
"epoch": 16.8,
"learning_rate": 8.000000000000001e-06,
"loss": 1.833,
"step": 210000
},
{
"epoch": 17.0,
"learning_rate": 7.5e-06,
"loss": 1.8452,
"step": 212500
},
{
"epoch": 17.2,
"learning_rate": 7.000000000000001e-06,
"loss": 1.6738,
"step": 215000
},
{
"epoch": 17.4,
"learning_rate": 6.5000000000000004e-06,
"loss": 1.6741,
"step": 217500
},
{
"epoch": 17.6,
"learning_rate": 6e-06,
"loss": 1.6924,
"step": 220000
},
{
"epoch": 17.8,
"learning_rate": 5.500000000000001e-06,
"loss": 1.6768,
"step": 222500
},
{
"epoch": 18.0,
"learning_rate": 5e-06,
"loss": 1.6922,
"step": 225000
},
{
"epoch": 18.2,
"learning_rate": 4.5e-06,
"loss": 1.567,
"step": 227500
},
{
"epoch": 18.4,
"learning_rate": 4.000000000000001e-06,
"loss": 1.5823,
"step": 230000
},
{
"epoch": 18.6,
"learning_rate": 3.5000000000000004e-06,
"loss": 1.6003,
"step": 232500
},
{
"epoch": 18.8,
"learning_rate": 3e-06,
"loss": 1.5739,
"step": 235000
},
{
"epoch": 19.0,
"learning_rate": 2.5e-06,
"loss": 1.5736,
"step": 237500
},
{
"epoch": 19.2,
"learning_rate": 2.0000000000000003e-06,
"loss": 1.5184,
"step": 240000
},
{
"epoch": 19.4,
"learning_rate": 1.5e-06,
"loss": 1.5043,
"step": 242500
},
{
"epoch": 19.6,
"learning_rate": 1.0000000000000002e-06,
"loss": 1.4998,
"step": 245000
},
{
"epoch": 19.8,
"learning_rate": 5.000000000000001e-07,
"loss": 1.4948,
"step": 247500
},
{
"epoch": 20.0,
"learning_rate": 0.0,
"loss": 1.5208,
"step": 250000
},
{
"epoch": 20.0,
"step": 250000,
"train_runtime": 72937.5039,
"train_samples_per_second": 3.428
}
],
"max_steps": 250000,
"num_train_epochs": 20,
"total_flos": 304359320699884944,
"trial_name": null,
"trial_params": null
}