inquisitive1 / trainer_state.json
kcarnold's picture
Import
b958694
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 7.0,
"global_step": 13944,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.25,
"learning_rate": 4.8207114170969595e-05,
"loss": 2.3679,
"step": 500
},
{
"epoch": 0.5,
"learning_rate": 4.641422834193919e-05,
"loss": 2.2289,
"step": 1000
},
{
"epoch": 0.75,
"learning_rate": 4.462134251290878e-05,
"loss": 2.1641,
"step": 1500
},
{
"epoch": 1.0,
"learning_rate": 4.2828456683878374e-05,
"loss": 2.1632,
"step": 2000
},
{
"epoch": 1.26,
"learning_rate": 4.1035570854847967e-05,
"loss": 1.7626,
"step": 2500
},
{
"epoch": 1.51,
"learning_rate": 3.924268502581756e-05,
"loss": 1.7654,
"step": 3000
},
{
"epoch": 1.76,
"learning_rate": 3.744979919678715e-05,
"loss": 1.7738,
"step": 3500
},
{
"epoch": 2.01,
"learning_rate": 3.5656913367756745e-05,
"loss": 1.7726,
"step": 4000
},
{
"epoch": 2.26,
"learning_rate": 3.386402753872633e-05,
"loss": 1.4475,
"step": 4500
},
{
"epoch": 2.51,
"learning_rate": 3.207114170969593e-05,
"loss": 1.465,
"step": 5000
},
{
"epoch": 2.76,
"learning_rate": 3.0278255880665517e-05,
"loss": 1.4945,
"step": 5500
},
{
"epoch": 3.01,
"learning_rate": 2.8485370051635113e-05,
"loss": 1.4736,
"step": 6000
},
{
"epoch": 3.26,
"learning_rate": 2.669248422260471e-05,
"loss": 1.2188,
"step": 6500
},
{
"epoch": 3.51,
"learning_rate": 2.48995983935743e-05,
"loss": 1.2392,
"step": 7000
},
{
"epoch": 3.77,
"learning_rate": 2.310671256454389e-05,
"loss": 1.244,
"step": 7500
},
{
"epoch": 4.02,
"learning_rate": 2.1313826735513484e-05,
"loss": 1.2268,
"step": 8000
},
{
"epoch": 4.27,
"learning_rate": 1.9520940906483077e-05,
"loss": 1.0412,
"step": 8500
},
{
"epoch": 4.52,
"learning_rate": 1.772805507745267e-05,
"loss": 1.0629,
"step": 9000
},
{
"epoch": 4.77,
"learning_rate": 1.593516924842226e-05,
"loss": 1.0673,
"step": 9500
},
{
"epoch": 5.02,
"learning_rate": 1.4142283419391852e-05,
"loss": 1.0521,
"step": 10000
},
{
"epoch": 5.27,
"learning_rate": 1.2349397590361447e-05,
"loss": 0.9034,
"step": 10500
},
{
"epoch": 5.52,
"learning_rate": 1.055651176133104e-05,
"loss": 0.9072,
"step": 11000
},
{
"epoch": 5.77,
"learning_rate": 8.763625932300632e-06,
"loss": 0.9298,
"step": 11500
},
{
"epoch": 6.02,
"learning_rate": 6.970740103270223e-06,
"loss": 0.9202,
"step": 12000
},
{
"epoch": 6.28,
"learning_rate": 5.177854274239816e-06,
"loss": 0.8249,
"step": 12500
},
{
"epoch": 6.53,
"learning_rate": 3.384968445209409e-06,
"loss": 0.8267,
"step": 13000
},
{
"epoch": 6.78,
"learning_rate": 1.5920826161790018e-06,
"loss": 0.8203,
"step": 13500
},
{
"epoch": 7.0,
"step": 13944,
"total_flos": 3714812226478080.0,
"train_loss": 1.3588120859379202,
"train_runtime": 1049.6357,
"train_samples_per_second": 106.244,
"train_steps_per_second": 13.285
}
],
"max_steps": 13944,
"num_train_epochs": 7,
"total_flos": 3714812226478080.0,
"trial_name": null,
"trial_params": null
}