checkpoint-375/trainer_state.json

{
  "best_metric": 0.8111211061477661,
  "best_model_checkpoint": "autotrain-paligemma-finetuned-vqa/checkpoint-375",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 375,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "grad_norm": 5.544835567474365,
      "learning_rate": 1.3157894736842108e-05,
      "loss": 1.6534,
      "step": 25
    },
    {
      "epoch": 0.4,
      "grad_norm": 6.025909423828125,
      "learning_rate": 1.9287833827893176e-05,
      "loss": 1.6205,
      "step": 50
    },
    {
      "epoch": 0.6,
      "grad_norm": 8.462043762207031,
      "learning_rate": 1.7804154302670624e-05,
      "loss": 1.1119,
      "step": 75
    },
    {
      "epoch": 0.8,
      "grad_norm": 3.2696471214294434,
      "learning_rate": 1.6320474777448072e-05,
      "loss": 0.7722,
      "step": 100
    },
    {
      "epoch": 1.0,
      "grad_norm": 5.208315372467041,
      "learning_rate": 1.483679525222552e-05,
      "loss": 0.8746,
      "step": 125
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.912371039390564,
      "eval_runtime": 24.3232,
      "eval_samples_per_second": 10.278,
      "eval_steps_per_second": 5.139,
      "step": 125
    },
    {
      "epoch": 1.2,
      "grad_norm": 3.5270237922668457,
      "learning_rate": 1.3353115727002968e-05,
      "loss": 0.6841,
      "step": 150
    },
    {
      "epoch": 1.4,
      "grad_norm": 3.0572621822357178,
      "learning_rate": 1.1869436201780416e-05,
      "loss": 0.7057,
      "step": 175
    },
    {
      "epoch": 1.6,
      "grad_norm": 5.3405561447143555,
      "learning_rate": 1.0385756676557866e-05,
      "loss": 0.6459,
      "step": 200
    },
    {
      "epoch": 1.8,
      "grad_norm": 4.48429012298584,
      "learning_rate": 8.902077151335312e-06,
      "loss": 0.6645,
      "step": 225
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.817482948303223,
      "learning_rate": 7.41839762611276e-06,
      "loss": 0.6855,
      "step": 250
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.827422022819519,
      "eval_runtime": 24.2956,
      "eval_samples_per_second": 10.29,
      "eval_steps_per_second": 5.145,
      "step": 250
    },
    {
      "epoch": 2.2,
      "grad_norm": 3.3924331665039062,
      "learning_rate": 5.934718100890208e-06,
      "loss": 0.5572,
      "step": 275
    },
    {
      "epoch": 2.4,
      "grad_norm": 8.049240112304688,
      "learning_rate": 4.451038575667656e-06,
      "loss": 0.5909,
      "step": 300
    },
    {
      "epoch": 2.6,
      "grad_norm": 3.2632410526275635,
      "learning_rate": 2.967359050445104e-06,
      "loss": 0.6251,
      "step": 325
    },
    {
      "epoch": 2.8,
      "grad_norm": 6.481785297393799,
      "learning_rate": 1.483679525222552e-06,
      "loss": 0.4504,
      "step": 350
    },
    {
      "epoch": 3.0,
      "grad_norm": 6.444954872131348,
      "learning_rate": 5.934718100890208e-08,
      "loss": 0.6476,
      "step": 375
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.8111211061477661,
      "eval_runtime": 24.2844,
      "eval_samples_per_second": 10.295,
      "eval_steps_per_second": 5.147,
      "step": 375
    }
  ],
  "logging_steps": 25,
  "max_steps": 375,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 1.1793929907551424e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}