Files
gpt2-verification/trainer_state.json

282 lines
7.1 KiB
JSON
Raw Permalink Normal View History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 3477,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.08628127696289906,
"grad_norm": 1.6037696599960327,
"learning_rate": 4.857635893011217e-05,
"loss": 8.298780517578125,
"step": 100
},
{
"epoch": 0.1725625539257981,
"grad_norm": 1.7524189949035645,
"learning_rate": 4.7138337647397185e-05,
"loss": 7.087490844726562,
"step": 200
},
{
"epoch": 0.25884383088869717,
"grad_norm": 1.324486494064331,
"learning_rate": 4.57003163646822e-05,
"loss": 6.956798095703125,
"step": 300
},
{
"epoch": 0.3451251078515962,
"grad_norm": 1.4535937309265137,
"learning_rate": 4.426229508196721e-05,
"loss": 6.822440795898437,
"step": 400
},
{
"epoch": 0.4314063848144953,
"grad_norm": 1.327022910118103,
"learning_rate": 4.282427379925223e-05,
"loss": 6.734780883789062,
"step": 500
},
{
"epoch": 0.5176876617773943,
"grad_norm": 1.1214697360992432,
"learning_rate": 4.138625251653725e-05,
"loss": 6.682239379882812,
"step": 600
},
{
"epoch": 0.6039689387402933,
"grad_norm": 1.5475066900253296,
"learning_rate": 3.994823123382226e-05,
"loss": 6.600054321289062,
"step": 700
},
{
"epoch": 0.6902502157031924,
"grad_norm": 1.3128082752227783,
"learning_rate": 3.8510209951107276e-05,
"loss": 6.516415405273437,
"step": 800
},
{
"epoch": 0.7765314926660914,
"grad_norm": 1.6903055906295776,
"learning_rate": 3.707218866839229e-05,
"loss": 6.505012817382813,
"step": 900
},
{
"epoch": 0.8628127696289906,
"grad_norm": 1.616003394126892,
"learning_rate": 3.563416738567731e-05,
"loss": 6.397700805664062,
"step": 1000
},
{
"epoch": 0.9490940465918896,
"grad_norm": 1.7408578395843506,
"learning_rate": 3.4196146102962325e-05,
"loss": 6.376287841796875,
"step": 1100
},
{
"epoch": 1.0353753235547887,
"grad_norm": 1.6899269819259644,
"learning_rate": 3.2758124820247346e-05,
"loss": 6.3477630615234375,
"step": 1200
},
{
"epoch": 1.1216566005176876,
"grad_norm": 1.5354853868484497,
"learning_rate": 3.132010353753236e-05,
"loss": 6.22337158203125,
"step": 1300
},
{
"epoch": 1.2079378774805867,
"grad_norm": 1.530237078666687,
"learning_rate": 2.9882082254817374e-05,
"loss": 6.231620483398437,
"step": 1400
},
{
"epoch": 1.2942191544434858,
"grad_norm": 1.71237313747406,
"learning_rate": 2.8444060972102388e-05,
"loss": 6.191749877929688,
"step": 1500
},
{
"epoch": 1.380500431406385,
"grad_norm": 1.532090425491333,
"learning_rate": 2.7006039689387402e-05,
"loss": 6.148622436523437,
"step": 1600
},
{
"epoch": 1.4667817083692838,
"grad_norm": 1.5732359886169434,
"learning_rate": 2.5568018406672423e-05,
"loss": 6.130806274414063,
"step": 1700
},
{
"epoch": 1.553062985332183,
"grad_norm": 1.74668288230896,
"learning_rate": 2.4129997123957434e-05,
"loss": 6.11656005859375,
"step": 1800
},
{
"epoch": 1.639344262295082,
"grad_norm": 1.6298942565917969,
"learning_rate": 2.269197584124245e-05,
"loss": 6.120827026367188,
"step": 1900
},
{
"epoch": 1.725625539257981,
"grad_norm": 1.5918290615081787,
"learning_rate": 2.125395455852747e-05,
"loss": 6.088132934570313,
"step": 2000
},
{
"epoch": 1.8119068162208802,
"grad_norm": 1.4154537916183472,
"learning_rate": 1.9815933275812482e-05,
"loss": 6.068784790039063,
"step": 2100
},
{
"epoch": 1.8981880931837791,
"grad_norm": 1.5682926177978516,
"learning_rate": 1.83779119930975e-05,
"loss": 6.038920288085937,
"step": 2200
},
{
"epoch": 1.984469370146678,
"grad_norm": 1.5493305921554565,
"learning_rate": 1.6939890710382514e-05,
"loss": 6.069011840820313,
"step": 2300
},
{
"epoch": 2.0707506471095773,
"grad_norm": 1.7473928928375244,
"learning_rate": 1.550186942766753e-05,
"loss": 6.010381469726562,
"step": 2400
},
{
"epoch": 2.1570319240724762,
"grad_norm": 1.7771306037902832,
"learning_rate": 1.4063848144952545e-05,
"loss": 5.94528076171875,
"step": 2500
},
{
"epoch": 2.243313201035375,
"grad_norm": 1.8056012392044067,
"learning_rate": 1.2625826862237561e-05,
"loss": 5.949163208007812,
"step": 2600
},
{
"epoch": 2.3295944779982745,
"grad_norm": 1.547564148902893,
"learning_rate": 1.1187805579522577e-05,
"loss": 5.962681884765625,
"step": 2700
},
{
"epoch": 2.4158757549611733,
"grad_norm": 1.4664490222930908,
"learning_rate": 9.749784296807593e-06,
"loss": 5.9176702880859375,
"step": 2800
},
{
"epoch": 2.5021570319240727,
"grad_norm": 1.5322167873382568,
"learning_rate": 8.31176301409261e-06,
"loss": 5.913838500976563,
"step": 2900
},
{
"epoch": 2.5884383088869716,
"grad_norm": 1.51213538646698,
"learning_rate": 6.873741731377626e-06,
"loss": 5.912578735351563,
"step": 3000
},
{
"epoch": 2.6747195858498705,
"grad_norm": 1.7505192756652832,
"learning_rate": 5.435720448662641e-06,
"loss": 5.9252081298828125,
"step": 3100
},
{
"epoch": 2.76100086281277,
"grad_norm": 1.5077073574066162,
"learning_rate": 3.997699165947656e-06,
"loss": 5.895281982421875,
"step": 3200
},
{
"epoch": 2.8472821397756687,
"grad_norm": 1.7371774911880493,
"learning_rate": 2.559677883232672e-06,
"loss": 5.896738891601562,
"step": 3300
},
{
"epoch": 2.9335634167385676,
"grad_norm": 1.5654444694519043,
"learning_rate": 1.1216566005176878e-06,
"loss": 5.89673583984375,
"step": 3400
},
{
"epoch": 3.0,
"step": 3477,
"total_flos": 3634049581056000.0,
"train_loss": 6.285199090465582,
"train_runtime": 228.5387,
"train_samples_per_second": 30.428,
"train_steps_per_second": 15.214
}
],
"logging_steps": 100,
"max_steps": 3477,
"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": true
},
"attributes": {}
}
},
"total_flos": 3634049581056000.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}