MNLP_M3_mcqa_model_alex_lr5 / trainer_state.json
ellendagher's picture
Upload folder using huggingface_hub
ffd382b verified
{
"best_global_step": 1037,
"best_metric": 0.4461146593093872,
"best_model_checkpoint": "./mcqa_qwen3_letter_alex_lr5/checkpoint-1037",
"epoch": 0.9995180722891567,
"eval_steps": 500,
"global_step": 1037,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04819277108433735,
"grad_norm": 50.355064392089844,
"learning_rate": 2.1634615384615387e-06,
"loss": 0.7352,
"step": 50
},
{
"epoch": 0.0963855421686747,
"grad_norm": 61.209354400634766,
"learning_rate": 4.51923076923077e-06,
"loss": 0.644,
"step": 100
},
{
"epoch": 0.14457831325301204,
"grad_norm": 85.26641082763672,
"learning_rate": 4.7856377277599145e-06,
"loss": 0.5845,
"step": 150
},
{
"epoch": 0.1927710843373494,
"grad_norm": 61.49842834472656,
"learning_rate": 4.5176848874598075e-06,
"loss": 0.5694,
"step": 200
},
{
"epoch": 0.24096385542168675,
"grad_norm": 30.45182228088379,
"learning_rate": 4.2497320471597005e-06,
"loss": 0.6183,
"step": 250
},
{
"epoch": 0.2891566265060241,
"grad_norm": 83.19606018066406,
"learning_rate": 3.981779206859593e-06,
"loss": 0.5863,
"step": 300
},
{
"epoch": 0.3373493975903614,
"grad_norm": 34.25322341918945,
"learning_rate": 3.713826366559486e-06,
"loss": 0.5868,
"step": 350
},
{
"epoch": 0.3855421686746988,
"grad_norm": 98.11255645751953,
"learning_rate": 3.4458735262593782e-06,
"loss": 0.544,
"step": 400
},
{
"epoch": 0.43373493975903615,
"grad_norm": 44.13338088989258,
"learning_rate": 3.1779206859592717e-06,
"loss": 0.5344,
"step": 450
},
{
"epoch": 0.4819277108433735,
"grad_norm": 43.46006393432617,
"learning_rate": 2.909967845659164e-06,
"loss": 0.511,
"step": 500
},
{
"epoch": 0.5301204819277109,
"grad_norm": 47.03801727294922,
"learning_rate": 2.6420150053590572e-06,
"loss": 0.5753,
"step": 550
},
{
"epoch": 0.5783132530120482,
"grad_norm": 50.981266021728516,
"learning_rate": 2.37406216505895e-06,
"loss": 0.499,
"step": 600
},
{
"epoch": 0.6265060240963856,
"grad_norm": 33.0101318359375,
"learning_rate": 2.106109324758843e-06,
"loss": 0.5683,
"step": 650
},
{
"epoch": 0.6746987951807228,
"grad_norm": 63.27317428588867,
"learning_rate": 1.8381564844587354e-06,
"loss": 0.5464,
"step": 700
},
{
"epoch": 0.7228915662650602,
"grad_norm": 46.463626861572266,
"learning_rate": 1.5702036441586282e-06,
"loss": 0.4691,
"step": 750
},
{
"epoch": 0.7710843373493976,
"grad_norm": 61.4338493347168,
"learning_rate": 1.302250803858521e-06,
"loss": 0.4503,
"step": 800
},
{
"epoch": 0.8192771084337349,
"grad_norm": 65.20037841796875,
"learning_rate": 1.0342979635584138e-06,
"loss": 0.4901,
"step": 850
},
{
"epoch": 0.8674698795180723,
"grad_norm": 25.02268409729004,
"learning_rate": 7.663451232583066e-07,
"loss": 0.4645,
"step": 900
},
{
"epoch": 0.9156626506024096,
"grad_norm": 38.329383850097656,
"learning_rate": 4.983922829581994e-07,
"loss": 0.5101,
"step": 950
},
{
"epoch": 0.963855421686747,
"grad_norm": 65.19310760498047,
"learning_rate": 2.304394426580922e-07,
"loss": 0.4518,
"step": 1000
},
{
"epoch": 0.9995180722891567,
"eval_loss": 0.4461146593093872,
"eval_runtime": 69.0209,
"eval_samples_per_second": 45.073,
"eval_steps_per_second": 5.636,
"step": 1037
}
],
"logging_steps": 50,
"max_steps": 1037,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"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": 7892894912348160.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}