EndoGPT / trainer_state.json
tombrooks248's picture
Upload V5 of the model
4f6ce4b
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.9625009745068995,
"global_step": 38000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 1.974013149346431e-05,
"loss": 0.3252,
"step": 500
},
{
"epoch": 0.08,
"learning_rate": 1.9480262986928614e-05,
"loss": 0.3841,
"step": 1000
},
{
"epoch": 0.12,
"learning_rate": 1.9220394480392923e-05,
"loss": 0.6714,
"step": 1500
},
{
"epoch": 0.16,
"learning_rate": 1.896052597385723e-05,
"loss": 0.7434,
"step": 2000
},
{
"epoch": 0.19,
"learning_rate": 1.8700657467321536e-05,
"loss": 0.3724,
"step": 2500
},
{
"epoch": 0.23,
"learning_rate": 1.8440788960785845e-05,
"loss": 0.4206,
"step": 3000
},
{
"epoch": 0.27,
"learning_rate": 1.818092045425015e-05,
"loss": 0.4183,
"step": 3500
},
{
"epoch": 0.31,
"learning_rate": 1.7921051947714458e-05,
"loss": 0.3506,
"step": 4000
},
{
"epoch": 0.35,
"learning_rate": 1.7661183441178764e-05,
"loss": 0.3634,
"step": 4500
},
{
"epoch": 0.39,
"learning_rate": 1.7401314934643073e-05,
"loss": 0.3824,
"step": 5000
},
{
"epoch": 0.43,
"learning_rate": 1.714144642810738e-05,
"loss": 0.4187,
"step": 5500
},
{
"epoch": 0.47,
"learning_rate": 1.6881577921571686e-05,
"loss": 0.3559,
"step": 6000
},
{
"epoch": 0.51,
"learning_rate": 1.6621709415035995e-05,
"loss": 0.3591,
"step": 6500
},
{
"epoch": 0.55,
"learning_rate": 1.6361840908500298e-05,
"loss": 0.3767,
"step": 7000
},
{
"epoch": 0.58,
"learning_rate": 1.6101972401964607e-05,
"loss": 0.3363,
"step": 7500
},
{
"epoch": 0.62,
"learning_rate": 1.5842103895428917e-05,
"loss": 0.3224,
"step": 8000
},
{
"epoch": 0.66,
"learning_rate": 1.558223538889322e-05,
"loss": 0.3362,
"step": 8500
},
{
"epoch": 0.7,
"learning_rate": 1.532236688235753e-05,
"loss": 0.3404,
"step": 9000
},
{
"epoch": 0.74,
"learning_rate": 1.5062498375821835e-05,
"loss": 0.3802,
"step": 9500
},
{
"epoch": 0.78,
"learning_rate": 1.4802629869286143e-05,
"loss": 0.3875,
"step": 10000
},
{
"epoch": 0.82,
"learning_rate": 1.4542761362750451e-05,
"loss": 0.4296,
"step": 10500
},
{
"epoch": 0.86,
"learning_rate": 1.4282892856214755e-05,
"loss": 0.3974,
"step": 11000
},
{
"epoch": 0.9,
"learning_rate": 1.4023024349679063e-05,
"loss": 0.3793,
"step": 11500
},
{
"epoch": 0.94,
"learning_rate": 1.376315584314337e-05,
"loss": 0.3472,
"step": 12000
},
{
"epoch": 0.97,
"learning_rate": 1.3503287336607677e-05,
"loss": 0.3598,
"step": 12500
},
{
"epoch": 1.0,
"eval_loss": 0.3997747600078583,
"eval_runtime": 662.7743,
"eval_samples_per_second": 66.354,
"eval_steps_per_second": 8.295,
"step": 12827
},
{
"epoch": 1.01,
"learning_rate": 1.3243418830071985e-05,
"loss": 0.3853,
"step": 13000
},
{
"epoch": 1.05,
"learning_rate": 1.2983550323536291e-05,
"loss": 0.3794,
"step": 13500
},
{
"epoch": 1.09,
"learning_rate": 1.2723681817000599e-05,
"loss": 0.3399,
"step": 14000
},
{
"epoch": 1.13,
"learning_rate": 1.2463813310464905e-05,
"loss": 0.3268,
"step": 14500
},
{
"epoch": 1.17,
"learning_rate": 1.2203944803929213e-05,
"loss": 0.3358,
"step": 15000
},
{
"epoch": 1.21,
"learning_rate": 1.1944076297393521e-05,
"loss": 0.3554,
"step": 15500
},
{
"epoch": 1.25,
"learning_rate": 1.1684207790857827e-05,
"loss": 0.337,
"step": 16000
},
{
"epoch": 1.29,
"learning_rate": 1.1424339284322135e-05,
"loss": 0.3321,
"step": 16500
},
{
"epoch": 1.33,
"learning_rate": 1.1164470777786441e-05,
"loss": 0.3314,
"step": 17000
},
{
"epoch": 1.36,
"learning_rate": 1.0904602271250749e-05,
"loss": 0.3243,
"step": 17500
},
{
"epoch": 1.4,
"learning_rate": 1.0644733764715057e-05,
"loss": 0.3147,
"step": 18000
},
{
"epoch": 1.44,
"learning_rate": 1.0384865258179361e-05,
"loss": 0.3184,
"step": 18500
},
{
"epoch": 1.48,
"learning_rate": 1.0124996751643669e-05,
"loss": 0.3638,
"step": 19000
},
{
"epoch": 1.52,
"learning_rate": 9.865128245107977e-06,
"loss": 0.3502,
"step": 19500
},
{
"epoch": 1.56,
"learning_rate": 9.605259738572283e-06,
"loss": 0.3486,
"step": 20000
},
{
"epoch": 1.6,
"learning_rate": 9.345391232036591e-06,
"loss": 0.388,
"step": 20500
},
{
"epoch": 1.64,
"learning_rate": 9.085522725500897e-06,
"loss": 0.4242,
"step": 21000
},
{
"epoch": 1.68,
"learning_rate": 8.825654218965203e-06,
"loss": 0.4053,
"step": 21500
},
{
"epoch": 1.72,
"learning_rate": 8.565785712429513e-06,
"loss": 0.3648,
"step": 22000
},
{
"epoch": 1.75,
"learning_rate": 8.305917205893819e-06,
"loss": 0.3675,
"step": 22500
},
{
"epoch": 1.79,
"learning_rate": 8.046048699358125e-06,
"loss": 0.3882,
"step": 23000
},
{
"epoch": 1.83,
"learning_rate": 7.786180192822433e-06,
"loss": 0.4318,
"step": 23500
},
{
"epoch": 1.87,
"learning_rate": 7.526311686286739e-06,
"loss": 0.402,
"step": 24000
},
{
"epoch": 1.91,
"learning_rate": 7.266443179751047e-06,
"loss": 0.3708,
"step": 24500
},
{
"epoch": 1.95,
"learning_rate": 7.006574673215354e-06,
"loss": 0.3576,
"step": 25000
},
{
"epoch": 1.99,
"learning_rate": 6.746706166679661e-06,
"loss": 0.3417,
"step": 25500
},
{
"epoch": 2.0,
"eval_loss": 0.3679686188697815,
"eval_runtime": 658.8287,
"eval_samples_per_second": 66.752,
"eval_steps_per_second": 8.345,
"step": 25654
},
{
"epoch": 2.03,
"learning_rate": 6.486837660143968e-06,
"loss": 0.355,
"step": 26000
},
{
"epoch": 2.07,
"learning_rate": 6.226969153608275e-06,
"loss": 0.3088,
"step": 26500
},
{
"epoch": 2.1,
"learning_rate": 5.967100647072583e-06,
"loss": 0.296,
"step": 27000
},
{
"epoch": 2.14,
"learning_rate": 5.707232140536889e-06,
"loss": 0.2952,
"step": 27500
},
{
"epoch": 2.18,
"learning_rate": 5.447363634001196e-06,
"loss": 0.2896,
"step": 28000
},
{
"epoch": 2.22,
"learning_rate": 5.187495127465503e-06,
"loss": 0.2768,
"step": 28500
},
{
"epoch": 2.26,
"learning_rate": 4.9276266209298106e-06,
"loss": 0.266,
"step": 29000
},
{
"epoch": 2.3,
"learning_rate": 4.667758114394117e-06,
"loss": 0.2753,
"step": 29500
},
{
"epoch": 2.34,
"learning_rate": 4.407889607858424e-06,
"loss": 0.2773,
"step": 30000
},
{
"epoch": 2.38,
"learning_rate": 4.1480211013227315e-06,
"loss": 0.2755,
"step": 30500
},
{
"epoch": 2.42,
"learning_rate": 3.888152594787038e-06,
"loss": 0.27,
"step": 31000
},
{
"epoch": 2.46,
"learning_rate": 3.6282840882513455e-06,
"loss": 0.2683,
"step": 31500
},
{
"epoch": 2.49,
"learning_rate": 3.368415581715652e-06,
"loss": 0.2718,
"step": 32000
},
{
"epoch": 2.53,
"learning_rate": 3.108547075179959e-06,
"loss": 0.2824,
"step": 32500
},
{
"epoch": 2.57,
"learning_rate": 2.8486785686442665e-06,
"loss": 0.2691,
"step": 33000
},
{
"epoch": 2.61,
"learning_rate": 2.588810062108573e-06,
"loss": 0.2664,
"step": 33500
},
{
"epoch": 2.65,
"learning_rate": 2.3289415555728805e-06,
"loss": 0.2655,
"step": 34000
},
{
"epoch": 2.69,
"learning_rate": 2.0690730490371875e-06,
"loss": 0.2681,
"step": 34500
},
{
"epoch": 2.73,
"learning_rate": 1.8092045425014945e-06,
"loss": 0.2751,
"step": 35000
},
{
"epoch": 2.77,
"learning_rate": 1.5493360359658013e-06,
"loss": 0.2731,
"step": 35500
},
{
"epoch": 2.81,
"learning_rate": 1.2894675294301085e-06,
"loss": 0.2642,
"step": 36000
},
{
"epoch": 2.85,
"learning_rate": 1.0295990228944154e-06,
"loss": 0.2622,
"step": 36500
},
{
"epoch": 2.88,
"learning_rate": 7.697305163587225e-07,
"loss": 0.2716,
"step": 37000
},
{
"epoch": 2.92,
"learning_rate": 5.098620098230296e-07,
"loss": 0.2717,
"step": 37500
},
{
"epoch": 2.96,
"learning_rate": 2.4999350328733664e-07,
"loss": 0.2672,
"step": 38000
}
],
"max_steps": 38481,
"num_train_epochs": 3,
"total_flos": 3.5290626392064e+16,
"trial_name": null,
"trial_params": null
}