279 lines
6.5 KiB
JSON
279 lines
6.5 KiB
JSON
{
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.0,
|
|
"eval_steps": 500,
|
|
"global_step": 35,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 0.05714285714285714,
|
|
"grad_norm": 0.9830285693355629,
|
|
"learning_rate": 6.666666666666667e-06,
|
|
"loss": 0.4363,
|
|
"step": 1
|
|
},
|
|
{
|
|
"epoch": 0.11428571428571428,
|
|
"grad_norm": 0.929627216331077,
|
|
"learning_rate": 1.3333333333333333e-05,
|
|
"loss": 0.4403,
|
|
"step": 2
|
|
},
|
|
{
|
|
"epoch": 0.17142857142857143,
|
|
"grad_norm": 1.3095300323636245,
|
|
"learning_rate": 2e-05,
|
|
"loss": 0.4581,
|
|
"step": 3
|
|
},
|
|
{
|
|
"epoch": 0.22857142857142856,
|
|
"grad_norm": 0.5520690121304136,
|
|
"learning_rate": 2.6666666666666667e-05,
|
|
"loss": 0.3056,
|
|
"step": 4
|
|
},
|
|
{
|
|
"epoch": 0.2857142857142857,
|
|
"grad_norm": 0.6754227161092171,
|
|
"learning_rate": 3.3333333333333335e-05,
|
|
"loss": 0.3,
|
|
"step": 5
|
|
},
|
|
{
|
|
"epoch": 0.34285714285714286,
|
|
"grad_norm": 0.8665396110865601,
|
|
"learning_rate": 4e-05,
|
|
"loss": 0.2475,
|
|
"step": 6
|
|
},
|
|
{
|
|
"epoch": 0.4,
|
|
"grad_norm": 0.8246163243765027,
|
|
"learning_rate": 3.9951281005196486e-05,
|
|
"loss": 0.2421,
|
|
"step": 7
|
|
},
|
|
{
|
|
"epoch": 0.45714285714285713,
|
|
"grad_norm": 0.5367049427560698,
|
|
"learning_rate": 3.980536137483141e-05,
|
|
"loss": 0.1815,
|
|
"step": 8
|
|
},
|
|
{
|
|
"epoch": 0.5142857142857142,
|
|
"grad_norm": 0.8642699624751934,
|
|
"learning_rate": 3.9562952014676116e-05,
|
|
"loss": 0.2042,
|
|
"step": 9
|
|
},
|
|
{
|
|
"epoch": 0.5714285714285714,
|
|
"grad_norm": 0.7801792447780528,
|
|
"learning_rate": 3.922523391876638e-05,
|
|
"loss": 0.1351,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 0.6285714285714286,
|
|
"grad_norm": 0.4874553279500772,
|
|
"learning_rate": 3.879385241571817e-05,
|
|
"loss": 0.1399,
|
|
"step": 11
|
|
},
|
|
{
|
|
"epoch": 0.6857142857142857,
|
|
"grad_norm": 0.2904789469170586,
|
|
"learning_rate": 3.827090915285202e-05,
|
|
"loss": 0.1204,
|
|
"step": 12
|
|
},
|
|
{
|
|
"epoch": 0.7428571428571429,
|
|
"grad_norm": 0.17567254221830908,
|
|
"learning_rate": 3.7658951857178544e-05,
|
|
"loss": 0.0933,
|
|
"step": 13
|
|
},
|
|
{
|
|
"epoch": 0.8,
|
|
"grad_norm": 0.28546802072873984,
|
|
"learning_rate": 3.696096192312852e-05,
|
|
"loss": 0.1165,
|
|
"step": 14
|
|
},
|
|
{
|
|
"epoch": 0.8571428571428571,
|
|
"grad_norm": 0.18853127265171502,
|
|
"learning_rate": 3.6180339887498953e-05,
|
|
"loss": 0.1052,
|
|
"step": 15
|
|
},
|
|
{
|
|
"epoch": 0.9142857142857143,
|
|
"grad_norm": 0.2069025715165584,
|
|
"learning_rate": 3.532088886237956e-05,
|
|
"loss": 0.0937,
|
|
"step": 16
|
|
},
|
|
{
|
|
"epoch": 0.9714285714285714,
|
|
"grad_norm": 0.4023933225317112,
|
|
"learning_rate": 3.438679600677303e-05,
|
|
"loss": 0.0879,
|
|
"step": 17
|
|
},
|
|
{
|
|
"epoch": 1.0285714285714285,
|
|
"grad_norm": 0.27482437218078265,
|
|
"learning_rate": 3.3382612127177166e-05,
|
|
"loss": 0.0988,
|
|
"step": 18
|
|
},
|
|
{
|
|
"epoch": 1.0857142857142856,
|
|
"grad_norm": 0.15942951571442726,
|
|
"learning_rate": 3.2313229506513167e-05,
|
|
"loss": 0.0656,
|
|
"step": 19
|
|
},
|
|
{
|
|
"epoch": 1.1428571428571428,
|
|
"grad_norm": 0.11238347551248191,
|
|
"learning_rate": 3.1183858069414936e-05,
|
|
"loss": 0.0582,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 1.2,
|
|
"grad_norm": 0.1480561246701996,
|
|
"learning_rate": 3.0000000000000004e-05,
|
|
"loss": 0.0619,
|
|
"step": 21
|
|
},
|
|
{
|
|
"epoch": 1.2571428571428571,
|
|
"grad_norm": 0.11327019366884013,
|
|
"learning_rate": 2.876742293578155e-05,
|
|
"loss": 0.0602,
|
|
"step": 22
|
|
},
|
|
{
|
|
"epoch": 1.3142857142857143,
|
|
"grad_norm": 0.1337906768933213,
|
|
"learning_rate": 2.7492131868318247e-05,
|
|
"loss": 0.0578,
|
|
"step": 23
|
|
},
|
|
{
|
|
"epoch": 1.3714285714285714,
|
|
"grad_norm": 0.14025183861943888,
|
|
"learning_rate": 2.618033988749895e-05,
|
|
"loss": 0.0503,
|
|
"step": 24
|
|
},
|
|
{
|
|
"epoch": 1.4285714285714286,
|
|
"grad_norm": 0.11330881808592926,
|
|
"learning_rate": 2.4838437911993355e-05,
|
|
"loss": 0.0534,
|
|
"step": 25
|
|
},
|
|
{
|
|
"epoch": 1.4857142857142858,
|
|
"grad_norm": 0.10689291386760147,
|
|
"learning_rate": 2.3472963553338614e-05,
|
|
"loss": 0.054,
|
|
"step": 26
|
|
},
|
|
{
|
|
"epoch": 1.5428571428571427,
|
|
"grad_norm": 0.12276021823647351,
|
|
"learning_rate": 2.209056926535307e-05,
|
|
"loss": 0.0555,
|
|
"step": 27
|
|
},
|
|
{
|
|
"epoch": 1.6,
|
|
"grad_norm": 0.1126246918210881,
|
|
"learning_rate": 2.0697989934050025e-05,
|
|
"loss": 0.0576,
|
|
"step": 28
|
|
},
|
|
{
|
|
"epoch": 1.657142857142857,
|
|
"grad_norm": 0.10819382916344927,
|
|
"learning_rate": 1.930201006594999e-05,
|
|
"loss": 0.0516,
|
|
"step": 29
|
|
},
|
|
{
|
|
"epoch": 1.7142857142857144,
|
|
"grad_norm": 0.10334535731906243,
|
|
"learning_rate": 1.7909430734646936e-05,
|
|
"loss": 0.0494,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 1.7714285714285714,
|
|
"grad_norm": 0.0997591568925581,
|
|
"learning_rate": 1.6527036446661396e-05,
|
|
"loss": 0.0491,
|
|
"step": 31
|
|
},
|
|
{
|
|
"epoch": 1.8285714285714287,
|
|
"grad_norm": 0.1065973125434392,
|
|
"learning_rate": 1.5161562088006649e-05,
|
|
"loss": 0.0477,
|
|
"step": 32
|
|
},
|
|
{
|
|
"epoch": 1.8857142857142857,
|
|
"grad_norm": 0.10230936923661474,
|
|
"learning_rate": 1.3819660112501054e-05,
|
|
"loss": 0.0477,
|
|
"step": 33
|
|
},
|
|
{
|
|
"epoch": 1.9428571428571428,
|
|
"grad_norm": 0.1182655999763823,
|
|
"learning_rate": 1.250786813168176e-05,
|
|
"loss": 0.0478,
|
|
"step": 34
|
|
},
|
|
{
|
|
"epoch": 2.0,
|
|
"grad_norm": 0.23512510361638483,
|
|
"learning_rate": 1.123257706421845e-05,
|
|
"loss": 0.0544,
|
|
"step": 35
|
|
}
|
|
],
|
|
"logging_steps": 1.0,
|
|
"max_steps": 51,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 3,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 91239600881664.0,
|
|
"train_batch_size": 1,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|