easyrec-small / trainer_state.json
jibala-1022's picture
Upload small model
25ec3bf verified
{
"best_metric": 0.01243716301208047,
"best_model_checkpoint": "./checkpoints/easyrec-small",
"epoch": 1.733531451213472,
"eval_steps": 1000,
"global_step": 14000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.06191183754333829,
"grad_norm": 24.75,
"learning_rate": 4.896813604094436e-05,
"loss": 3.2576,
"step": 500
},
{
"epoch": 0.12382367508667658,
"grad_norm": 21.75,
"learning_rate": 4.793627208188873e-05,
"loss": 2.7834,
"step": 1000
},
{
"epoch": 0.18573551263001487,
"grad_norm": 21.25,
"learning_rate": 4.6904408122833086e-05,
"loss": 2.6955,
"step": 1500
},
{
"epoch": 0.24764735017335315,
"grad_norm": 24.375,
"learning_rate": 4.587254416377745e-05,
"loss": 2.5928,
"step": 2000
},
{
"epoch": 0.30955918771669144,
"grad_norm": 27.875,
"learning_rate": 4.484068020472182e-05,
"loss": 2.5848,
"step": 2500
},
{
"epoch": 0.37147102526002973,
"grad_norm": 21.625,
"learning_rate": 4.3808816245666175e-05,
"loss": 2.5582,
"step": 3000
},
{
"epoch": 0.433382862803368,
"grad_norm": 27.125,
"learning_rate": 4.2776952286610534e-05,
"loss": 2.5386,
"step": 3500
},
{
"epoch": 0.4952947003467063,
"grad_norm": 29.625,
"learning_rate": 4.17450883275549e-05,
"loss": 2.5143,
"step": 4000
},
{
"epoch": 0.5572065378900446,
"grad_norm": 20.0,
"learning_rate": 4.071322436849926e-05,
"loss": 2.5071,
"step": 4500
},
{
"epoch": 0.6191183754333829,
"grad_norm": 26.25,
"learning_rate": 3.968136040944362e-05,
"loss": 2.4754,
"step": 5000
},
{
"epoch": 0.6810302129767212,
"grad_norm": 23.0,
"learning_rate": 3.864949645038798e-05,
"loss": 2.4846,
"step": 5500
},
{
"epoch": 0.7429420505200595,
"grad_norm": 24.5,
"learning_rate": 3.761763249133234e-05,
"loss": 2.4806,
"step": 6000
},
{
"epoch": 0.8048538880633977,
"grad_norm": 24.125,
"learning_rate": 3.658576853227671e-05,
"loss": 2.4532,
"step": 6500
},
{
"epoch": 0.866765725606736,
"grad_norm": 24.25,
"learning_rate": 3.555390457322107e-05,
"loss": 2.4561,
"step": 7000
},
{
"epoch": 0.9286775631500743,
"grad_norm": 30.75,
"learning_rate": 3.452204061416543e-05,
"loss": 2.4607,
"step": 7500
},
{
"epoch": 0.9905894006934126,
"grad_norm": 26.375,
"learning_rate": 3.349017665510979e-05,
"loss": 2.4332,
"step": 8000
},
{
"epoch": 1.052501238236751,
"grad_norm": 29.375,
"learning_rate": 3.2458312696054156e-05,
"loss": 2.4328,
"step": 8500
},
{
"epoch": 1.1144130757800892,
"grad_norm": 25.0,
"learning_rate": 3.1426448736998515e-05,
"loss": 2.4362,
"step": 9000
},
{
"epoch": 1.1763249133234275,
"grad_norm": 24.25,
"learning_rate": 3.0394584777942874e-05,
"loss": 2.4432,
"step": 9500
},
{
"epoch": 1.2382367508667658,
"grad_norm": 34.75,
"learning_rate": 2.936272081888724e-05,
"loss": 2.427,
"step": 10000
},
{
"epoch": 1.300148588410104,
"grad_norm": 24.5,
"learning_rate": 2.83308568598316e-05,
"loss": 2.4219,
"step": 10500
},
{
"epoch": 1.3620604259534423,
"grad_norm": 35.0,
"learning_rate": 2.729899290077596e-05,
"loss": 2.4213,
"step": 11000
},
{
"epoch": 1.4239722634967806,
"grad_norm": 28.75,
"learning_rate": 2.6267128941720326e-05,
"loss": 2.424,
"step": 11500
},
{
"epoch": 1.485884101040119,
"grad_norm": 26.0,
"learning_rate": 2.5235264982664684e-05,
"loss": 2.4192,
"step": 12000
},
{
"epoch": 1.5477959385834572,
"grad_norm": 22.75,
"learning_rate": 2.420340102360905e-05,
"loss": 2.4032,
"step": 12500
},
{
"epoch": 1.6097077761267955,
"grad_norm": 35.0,
"learning_rate": 2.3171537064553412e-05,
"loss": 2.4317,
"step": 13000
},
{
"epoch": 1.6716196136701336,
"grad_norm": 20.875,
"learning_rate": 2.213967310549777e-05,
"loss": 2.432,
"step": 13500
},
{
"epoch": 1.733531451213472,
"grad_norm": 25.625,
"learning_rate": 2.1107809146442133e-05,
"loss": 2.4309,
"step": 14000
}
],
"logging_steps": 500,
"max_steps": 24228,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 1000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": false,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 0.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}