falcon-7b-sft-lora / trainer_state.json
Mastane's picture
Model save
bbf3802
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9852216748768473,
"eval_steps": 500,
"global_step": 50,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 1.9980267284282718e-05,
"loss": 1.7982,
"step": 1
},
{
"epoch": 0.1,
"learning_rate": 1.9510565162951538e-05,
"loss": 1.7845,
"step": 5
},
{
"epoch": 0.2,
"learning_rate": 1.8090169943749477e-05,
"loss": 1.7893,
"step": 10
},
{
"epoch": 0.3,
"learning_rate": 1.5877852522924733e-05,
"loss": 1.768,
"step": 15
},
{
"epoch": 0.39,
"learning_rate": 1.3090169943749475e-05,
"loss": 1.749,
"step": 20
},
{
"epoch": 0.49,
"learning_rate": 1e-05,
"loss": 1.7308,
"step": 25
},
{
"epoch": 0.59,
"learning_rate": 6.909830056250527e-06,
"loss": 1.7066,
"step": 30
},
{
"epoch": 0.69,
"learning_rate": 4.12214747707527e-06,
"loss": 1.6928,
"step": 35
},
{
"epoch": 0.79,
"learning_rate": 1.9098300562505266e-06,
"loss": 1.6886,
"step": 40
},
{
"epoch": 0.89,
"learning_rate": 4.894348370484648e-07,
"loss": 1.6787,
"step": 45
},
{
"epoch": 0.99,
"learning_rate": 0.0,
"loss": 1.6877,
"step": 50
},
{
"epoch": 0.99,
"eval_loss": 1.6897964477539062,
"eval_runtime": 1242.0222,
"eval_samples_per_second": 18.607,
"eval_steps_per_second": 0.291,
"step": 50
},
{
"epoch": 0.99,
"step": 50,
"total_flos": 4.2510105305184993e+18,
"train_loss": 1.7278787899017334,
"train_runtime": 16160.382,
"train_samples_per_second": 12.863,
"train_steps_per_second": 0.003
}
],
"logging_steps": 5,
"max_steps": 50,
"num_train_epochs": 1,
"save_steps": 500,
"total_flos": 4.2510105305184993e+18,
"trial_name": null,
"trial_params": null
}