breeze-listen-dsw-base-te / trainer_state.json
hanasim's picture
End of training
a3a68ac verified
raw
history blame
2.04 kB
{
"best_metric": 3.3333333333333335,
"best_model_checkpoint": "/cosmos/home/sp-operator/ai/training/models/huggingface/scripts/../breeze-listen-dsw-base-te/checkpoint-200",
"epoch": 99.001,
"eval_steps": 200,
"global_step": 200,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 12.0,
"learning_rate": 4.973833272194737e-06,
"loss": 1.7212,
"step": 25
},
{
"epoch": 24.0,
"learning_rate": 6.195318418690893e-06,
"loss": 1.1394,
"step": 50
},
{
"epoch": 37.0,
"learning_rate": 6.881634451095711e-06,
"loss": 0.5083,
"step": 75
},
{
"epoch": 49.0,
"learning_rate": 7.361221988663844e-06,
"loss": 0.1183,
"step": 100
},
{
"epoch": 62.0,
"learning_rate": 7.730207550743121e-06,
"loss": 0.022,
"step": 125
},
{
"epoch": 74.0,
"learning_rate": 8.03016458599496e-06,
"loss": 0.005,
"step": 150
},
{
"epoch": 87.0,
"learning_rate": 8.282894746203441e-06,
"loss": 0.0016,
"step": 175
},
{
"epoch": 99.0,
"learning_rate": 8.501266121799902e-06,
"loss": 0.0008,
"step": 200
},
{
"epoch": 99.0,
"eval_loss": 0.05462646484375,
"eval_runtime": 53.8277,
"eval_samples_per_second": 0.725,
"eval_steps_per_second": 0.056,
"eval_wer": 3.3333333333333335,
"step": 200
},
{
"epoch": 99.0,
"step": 200,
"total_flos": 2.2700998962393907e+17,
"train_loss": 0.0,
"train_runtime": 2.6625,
"train_samples_per_second": 24037.247,
"train_steps_per_second": 751.164
}
],
"logging_steps": 25,
"max_steps": 2000,
"num_input_tokens_seen": 0,
"num_train_epochs": 9223372036854775807,
"save_steps": 200,
"total_flos": 2.2700998962393907e+17,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}