vit-base-beans-demo-v5 / trainer_state.json
Jipumpkin's picture
Jipympkin
3e4f373 verified
{
"best_metric": 0.01621263474225998,
"best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-200",
"epoch": 4.0,
"eval_steps": 100,
"global_step": 260,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.15384615384615385,
"grad_norm": 1.5753413438796997,
"learning_rate": 0.00019230769230769233,
"loss": 0.7775,
"step": 10
},
{
"epoch": 0.3076923076923077,
"grad_norm": 1.6680392026901245,
"learning_rate": 0.00018461538461538463,
"loss": 0.3173,
"step": 20
},
{
"epoch": 0.46153846153846156,
"grad_norm": 1.550700068473816,
"learning_rate": 0.00017692307692307693,
"loss": 0.1897,
"step": 30
},
{
"epoch": 0.6153846153846154,
"grad_norm": 1.188010811805725,
"learning_rate": 0.00016923076923076923,
"loss": 0.2146,
"step": 40
},
{
"epoch": 0.7692307692307693,
"grad_norm": 0.5321382284164429,
"learning_rate": 0.00016153846153846155,
"loss": 0.253,
"step": 50
},
{
"epoch": 0.9230769230769231,
"grad_norm": 1.6223677396774292,
"learning_rate": 0.00015384615384615385,
"loss": 0.2528,
"step": 60
},
{
"epoch": 1.0769230769230769,
"grad_norm": 0.35057711601257324,
"learning_rate": 0.00014615384615384615,
"loss": 0.1648,
"step": 70
},
{
"epoch": 1.2307692307692308,
"grad_norm": 0.20589804649353027,
"learning_rate": 0.00013846153846153847,
"loss": 0.113,
"step": 80
},
{
"epoch": 1.3846153846153846,
"grad_norm": 2.785675287246704,
"learning_rate": 0.00013076923076923077,
"loss": 0.0934,
"step": 90
},
{
"epoch": 1.5384615384615383,
"grad_norm": 0.8817322254180908,
"learning_rate": 0.0001230769230769231,
"loss": 0.1104,
"step": 100
},
{
"epoch": 1.5384615384615383,
"eval_accuracy": 0.9398496240601504,
"eval_loss": 0.17710085213184357,
"eval_runtime": 1.2887,
"eval_samples_per_second": 103.204,
"eval_steps_per_second": 13.191,
"step": 100
},
{
"epoch": 1.6923076923076923,
"grad_norm": 2.9832639694213867,
"learning_rate": 0.00011538461538461538,
"loss": 0.1418,
"step": 110
},
{
"epoch": 1.8461538461538463,
"grad_norm": 0.1858932077884674,
"learning_rate": 0.0001076923076923077,
"loss": 0.1041,
"step": 120
},
{
"epoch": 2.0,
"grad_norm": 12.24283218383789,
"learning_rate": 0.0001,
"loss": 0.0732,
"step": 130
},
{
"epoch": 2.1538461538461537,
"grad_norm": 0.11072707176208496,
"learning_rate": 9.230769230769232e-05,
"loss": 0.1019,
"step": 140
},
{
"epoch": 2.3076923076923075,
"grad_norm": 0.35365092754364014,
"learning_rate": 8.461538461538461e-05,
"loss": 0.0598,
"step": 150
},
{
"epoch": 2.4615384615384617,
"grad_norm": 3.38576340675354,
"learning_rate": 7.692307692307693e-05,
"loss": 0.0517,
"step": 160
},
{
"epoch": 2.6153846153846154,
"grad_norm": 0.34387993812561035,
"learning_rate": 6.923076923076924e-05,
"loss": 0.0736,
"step": 170
},
{
"epoch": 2.769230769230769,
"grad_norm": 0.09394571185112,
"learning_rate": 6.153846153846155e-05,
"loss": 0.0164,
"step": 180
},
{
"epoch": 2.9230769230769234,
"grad_norm": 0.09971573948860168,
"learning_rate": 5.384615384615385e-05,
"loss": 0.0289,
"step": 190
},
{
"epoch": 3.076923076923077,
"grad_norm": 0.07261139899492264,
"learning_rate": 4.615384615384616e-05,
"loss": 0.0152,
"step": 200
},
{
"epoch": 3.076923076923077,
"eval_accuracy": 1.0,
"eval_loss": 0.01621263474225998,
"eval_runtime": 1.2728,
"eval_samples_per_second": 104.497,
"eval_steps_per_second": 13.357,
"step": 200
},
{
"epoch": 3.230769230769231,
"grad_norm": 0.06883873790502548,
"learning_rate": 3.846153846153846e-05,
"loss": 0.0361,
"step": 210
},
{
"epoch": 3.3846153846153846,
"grad_norm": 0.06754733622074127,
"learning_rate": 3.0769230769230774e-05,
"loss": 0.013,
"step": 220
},
{
"epoch": 3.5384615384615383,
"grad_norm": 0.07049394398927689,
"learning_rate": 2.307692307692308e-05,
"loss": 0.0126,
"step": 230
},
{
"epoch": 3.6923076923076925,
"grad_norm": 0.06614019721746445,
"learning_rate": 1.5384615384615387e-05,
"loss": 0.0126,
"step": 240
},
{
"epoch": 3.8461538461538463,
"grad_norm": 0.08794592320919037,
"learning_rate": 7.692307692307694e-06,
"loss": 0.013,
"step": 250
},
{
"epoch": 4.0,
"grad_norm": 0.05894031003117561,
"learning_rate": 0.0,
"loss": 0.0119,
"step": 260
},
{
"epoch": 4.0,
"step": 260,
"total_flos": 3.205097416476426e+17,
"train_loss": 0.1250896595705014,
"train_runtime": 89.1507,
"train_samples_per_second": 46.393,
"train_steps_per_second": 2.916
}
],
"logging_steps": 10,
"max_steps": 260,
"num_input_tokens_seen": 0,
"num_train_epochs": 4,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3.205097416476426e+17,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}