trainer_state.json

{
  "best_metric": 0.06312847882509232,
  "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": 0.09027902781963348,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.0205,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 3.1298561096191406,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.0368,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.06700307130813599,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.0168,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.1318458616733551,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.109,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 3.4888665676116943,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.0709,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.1447253227233887,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.1581,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 3.413907527923584,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.1575,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 6.059014320373535,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.0474,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.14840804040431976,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0355,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 4.872044563293457,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.1086,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.09510985016822815,
      "eval_runtime": 1.2631,
      "eval_samples_per_second": 105.294,
      "eval_steps_per_second": 13.459,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 1.8685437440872192,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.0997,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.28649815917015076,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.1497,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.658267974853516,
      "learning_rate": 0.0001,
      "loss": 0.0504,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 2.9416344165802,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0137,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 7.906864166259766,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0201,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.22749103605747223,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.011,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.11756265163421631,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0328,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.0330984964966774,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0396,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.04163922742009163,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0354,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.03229311481118202,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0061,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.06312847882509232,
      "eval_runtime": 1.2372,
      "eval_samples_per_second": 107.501,
      "eval_steps_per_second": 13.741,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.054849058389663696,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.006,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.031628116965293884,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0059,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 9.332858085632324,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0113,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.030381686985492706,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0056,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.028255194425582886,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0054,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.027122126892209053,
      "learning_rate": 0.0,
      "loss": 0.0055,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.04844067517954569,
      "train_runtime": 89.4258,
      "train_samples_per_second": 46.251,
      "train_steps_per_second": 2.907
    }
  ],
  "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
}