TechTokenBert / trainer_state.json
AndreaTacchella's picture
Upload folder using huggingface_hub
5121458 verified
{
"best_global_step": 551560,
"best_metric": 0.4202389121055603,
"best_model_checkpoint": "submission/results/model_TT_7dcodes_cls_tt_end_not_shuffled/checkpoint-551560",
"epoch": 2.0,
"eval_steps": 275780,
"global_step": 551564,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.09999927478950765,
"grad_norm": 4.513119220733643,
"learning_rate": 4.7543207322561444e-05,
"loss": 0.7242,
"step": 27578
},
{
"epoch": 0.1999985495790153,
"grad_norm": 3.410264015197754,
"learning_rate": 4.50409571302063e-05,
"loss": 0.621,
"step": 55156
},
{
"epoch": 0.29999782436852296,
"grad_norm": 3.972125768661499,
"learning_rate": 4.253870693785114e-05,
"loss": 0.6056,
"step": 82734
},
{
"epoch": 0.3999970991580306,
"grad_norm": 3.088068962097168,
"learning_rate": 4.0036456745495986e-05,
"loss": 0.5958,
"step": 110312
},
{
"epoch": 0.49999637394753826,
"grad_norm": 2.8378355503082275,
"learning_rate": 3.753420655314083e-05,
"loss": 0.5842,
"step": 137890
},
{
"epoch": 0.5999956487370459,
"grad_norm": 2.7692346572875977,
"learning_rate": 3.503195636078569e-05,
"loss": 0.5735,
"step": 165468
},
{
"epoch": 0.6999949235265536,
"grad_norm": 4.5936598777771,
"learning_rate": 3.252970616843053e-05,
"loss": 0.5644,
"step": 193046
},
{
"epoch": 0.7999941983160612,
"grad_norm": 3.122809648513794,
"learning_rate": 3.0027455976075375e-05,
"loss": 0.5521,
"step": 220624
},
{
"epoch": 0.8999934731055689,
"grad_norm": 2.5449647903442383,
"learning_rate": 2.7525205783720222e-05,
"loss": 0.5396,
"step": 248202
},
{
"epoch": 0.9999927478950765,
"grad_norm": 2.232771635055542,
"learning_rate": 2.5022955591365072e-05,
"loss": 0.53,
"step": 275780
},
{
"epoch": 0.9999927478950765,
"eval_loss": 0.4933322072029114,
"eval_runtime": 6025.5826,
"eval_samples_per_second": 61.025,
"eval_steps_per_second": 5.085,
"step": 275780
},
{
"epoch": 1.0999920226845843,
"grad_norm": 1.9279263019561768,
"learning_rate": 2.2520705399009916e-05,
"loss": 0.5164,
"step": 303358
},
{
"epoch": 1.1999912974740918,
"grad_norm": 2.2434041500091553,
"learning_rate": 2.0018455206654763e-05,
"loss": 0.5062,
"step": 330936
},
{
"epoch": 1.2999905722635994,
"grad_norm": 4.116876602172852,
"learning_rate": 1.751620501429961e-05,
"loss": 0.4966,
"step": 358514
},
{
"epoch": 1.3999898470531071,
"grad_norm": 2.3324735164642334,
"learning_rate": 1.5013954821944456e-05,
"loss": 0.4867,
"step": 386092
},
{
"epoch": 1.499989121842615,
"grad_norm": 2.4795002937316895,
"learning_rate": 1.2511704629589305e-05,
"loss": 0.4773,
"step": 413670
},
{
"epoch": 1.5999883966321224,
"grad_norm": 2.851177453994751,
"learning_rate": 1.0009454437234152e-05,
"loss": 0.4695,
"step": 441248
},
{
"epoch": 1.69998767142163,
"grad_norm": 2.3028323650360107,
"learning_rate": 7.507204244878999e-06,
"loss": 0.4601,
"step": 468826
},
{
"epoch": 1.7999869462111378,
"grad_norm": 2.5855956077575684,
"learning_rate": 5.004954052523845e-06,
"loss": 0.4512,
"step": 496404
},
{
"epoch": 1.8999862210006455,
"grad_norm": 3.5067687034606934,
"learning_rate": 2.502703860168692e-06,
"loss": 0.4444,
"step": 523982
},
{
"epoch": 1.999985495790153,
"grad_norm": 3.1099815368652344,
"learning_rate": 4.536678135388993e-10,
"loss": 0.4395,
"step": 551560
},
{
"epoch": 1.999985495790153,
"eval_loss": 0.4202389121055603,
"eval_runtime": 6034.8002,
"eval_samples_per_second": 60.931,
"eval_steps_per_second": 5.078,
"step": 551560
}
],
"logging_steps": 27578,
"max_steps": 551564,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 27578,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 6.1691929035264e+18,
"train_batch_size": 12,
"trial_name": null,
"trial_params": null
}