joseph10's picture
Training in progress, epoch 2
e263a16
raw
history blame
1.35 kB
{
"best_metric": 0.8747714808043876,
"best_model_checkpoint": "berttiny-hate_speech18-bothpretrained\\run-17\\checkpoint-3830",
"epoch": 2.0,
"eval_steps": 500,
"global_step": 3830,
"is_hyper_param_search": true,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"learning_rate": 4.804034664994629e-05,
"loss": 0.4759,
"step": 1915
},
{
"epoch": 1.0,
"eval_accuracy": 0.863345521023766,
"eval_loss": 0.4774511456489563,
"eval_runtime": 10.6465,
"eval_samples_per_second": 205.513,
"eval_steps_per_second": 1.691,
"step": 1915
},
{
"epoch": 2.0,
"learning_rate": 0.0,
"loss": 0.4517,
"step": 3830
},
{
"epoch": 2.0,
"eval_accuracy": 0.8747714808043876,
"eval_loss": 0.4742494225502014,
"eval_runtime": 10.7351,
"eval_samples_per_second": 203.818,
"eval_steps_per_second": 1.677,
"step": 3830
}
],
"logging_steps": 500,
"max_steps": 3830,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 500,
"total_flos": 1545627716256.0,
"trial_name": null,
"trial_params": {
"learning_rate": 9.608069329989258e-05,
"num_train_epochs": 2,
"per_device_train_batch_size": 4,
"seed": 20
}
}