trainer_state.json

{
  "best_metric": 0.003522987710312009,
  "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": 0.04833316430449486,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.0099,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.04249303787946701,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.0084,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.03950580209493637,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.0074,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.038794852793216705,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.0067,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.03075353428721428,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.0061,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.028776561841368675,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.0057,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 0.02866475097835064,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.0052,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.02523769624531269,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.0048,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.02398413047194481,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0045,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.02361473999917507,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0043,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0047020018100738525,
      "eval_runtime": 1.6788,
      "eval_samples_per_second": 79.221,
      "eval_steps_per_second": 10.126,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 0.022088322788476944,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.0041,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.02115621790289879,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.0039,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.01992017589509487,
      "learning_rate": 0.0001,
      "loss": 0.0038,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.019350819289684296,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0036,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.019210871309041977,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0035,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.018969299271702766,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0034,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.018202364444732666,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0033,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.017885109409689903,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0032,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.017497321590781212,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0032,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.016919244080781937,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0031,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 1.0,
      "eval_loss": 0.003522987710312009,
      "eval_runtime": 1.2569,
      "eval_samples_per_second": 105.819,
      "eval_steps_per_second": 13.526,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.017208773642778397,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.003,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.016676107421517372,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.003,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.017159927636384964,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.003,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.0164987500756979,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0029,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.017116356641054153,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.003,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0159891489893198,
      "learning_rate": 0.0,
      "loss": 0.0029,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.004455769520539504,
      "train_runtime": 87.2945,
      "train_samples_per_second": 47.38,
      "train_steps_per_second": 2.978
    }
  ],
  "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
}