flan_t5_xl_cnn_dailymail / trainer_state.json
Ubuntu
First model version
9ceb86a
{
"best_metric": 1.2524826526641846,
"best_model_checkpoint": "flan-t5-xl/checkpoint-5982",
"epoch": 3.0,
"global_step": 8973,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.17,
"learning_rate": 0.0001,
"loss": 1.4987,
"step": 500
},
{
"epoch": 0.33,
"learning_rate": 0.0001,
"loss": 1.4692,
"step": 1000
},
{
"epoch": 0.5,
"learning_rate": 0.0001,
"loss": 1.4559,
"step": 1500
},
{
"epoch": 0.67,
"learning_rate": 0.0001,
"loss": 1.4418,
"step": 2000
},
{
"epoch": 0.84,
"learning_rate": 0.0001,
"loss": 1.4352,
"step": 2500
},
{
"epoch": 1.0,
"eval_gen_len": 101.99678336325553,
"eval_loss": 1.264474868774414,
"eval_rouge1": 43.8582,
"eval_rouge2": 21.2227,
"eval_rougeL": 30.7038,
"eval_rougeLsum": 40.761,
"eval_runtime": 23490.726,
"eval_samples_per_second": 0.569,
"eval_steps_per_second": 0.006,
"step": 2991
},
{
"epoch": 1.0,
"learning_rate": 0.0001,
"loss": 1.426,
"step": 3000
},
{
"epoch": 1.17,
"learning_rate": 0.0001,
"loss": 1.3106,
"step": 3500
},
{
"epoch": 1.34,
"learning_rate": 0.0001,
"loss": 1.3127,
"step": 4000
},
{
"epoch": 1.5,
"learning_rate": 0.0001,
"loss": 1.3173,
"step": 4500
},
{
"epoch": 1.67,
"learning_rate": 0.0001,
"loss": 1.3158,
"step": 5000
},
{
"epoch": 1.84,
"learning_rate": 0.0001,
"loss": 1.3198,
"step": 5500
},
{
"epoch": 2.0,
"eval_gen_len": 94.07330939557151,
"eval_loss": 1.2524826526641846,
"eval_rouge1": 44.4594,
"eval_rouge2": 21.8174,
"eval_rougeL": 31.4304,
"eval_rougeLsum": 41.4563,
"eval_runtime": 23248.3801,
"eval_samples_per_second": 0.575,
"eval_steps_per_second": 0.006,
"step": 5982
},
{
"epoch": 2.01,
"learning_rate": 0.0001,
"loss": 1.3143,
"step": 6000
},
{
"epoch": 2.17,
"learning_rate": 0.0001,
"loss": 1.1993,
"step": 6500
},
{
"epoch": 2.34,
"learning_rate": 0.0001,
"loss": 1.2054,
"step": 7000
},
{
"epoch": 2.51,
"learning_rate": 0.0001,
"loss": 1.2063,
"step": 7500
},
{
"epoch": 2.67,
"learning_rate": 0.0001,
"loss": 1.2095,
"step": 8000
},
{
"epoch": 2.84,
"learning_rate": 0.0001,
"loss": 1.2151,
"step": 8500
},
{
"epoch": 3.0,
"eval_gen_len": 94.23324356672651,
"eval_loss": 1.264769434928894,
"eval_rouge1": 45.1318,
"eval_rouge2": 22.2773,
"eval_rougeL": 31.9084,
"eval_rougeLsum": 42.0558,
"eval_runtime": 23125.233,
"eval_samples_per_second": 0.578,
"eval_steps_per_second": 0.006,
"step": 8973
},
{
"epoch": 3.0,
"step": 8973,
"total_flos": 650731472289792.0,
"train_loss": 1.326217149713028,
"train_runtime": 208860.2672,
"train_samples_per_second": 4.124,
"train_steps_per_second": 0.043
}
],
"max_steps": 8973,
"num_train_epochs": 3,
"total_flos": 650731472289792.0,
"trial_name": null,
"trial_params": null
}