checkpoint-3148/trainer_state.json

{
  "best_metric": 0.4061734974384308,
  "best_model_checkpoint": "facial_emotions_image_detection/checkpoint-3148",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 3148,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32,
      "grad_norm": 10.101922035217285,
      "learning_rate": 4.273724983860556e-06,
      "loss": 0.5503,
      "step": 500
    },
    {
      "epoch": 0.64,
      "grad_norm": 14.047916412353516,
      "learning_rate": 3.466752743705617e-06,
      "loss": 0.5278,
      "step": 1000
    },
    {
      "epoch": 0.95,
      "grad_norm": 10.702422142028809,
      "learning_rate": 2.6597805035506778e-06,
      "loss": 0.5172,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8700834326579261,
      "eval_loss": 0.41607460379600525,
      "eval_runtime": 111.1172,
      "eval_samples_per_second": 113.259,
      "eval_steps_per_second": 14.165,
      "step": 1574
    },
    {
      "epoch": 1.27,
      "grad_norm": 13.397717475891113,
      "learning_rate": 1.8528082633957392e-06,
      "loss": 0.4672,
      "step": 2000
    },
    {
      "epoch": 1.59,
      "grad_norm": 16.72222900390625,
      "learning_rate": 1.0458360232408006e-06,
      "loss": 0.4599,
      "step": 2500
    },
    {
      "epoch": 1.91,
      "grad_norm": 11.587815284729004,
      "learning_rate": 2.3886378308586184e-07,
      "loss": 0.4558,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8742947953913389,
      "eval_loss": 0.4061734974384308,
      "eval_runtime": 110.2844,
      "eval_samples_per_second": 114.114,
      "eval_steps_per_second": 14.272,
      "step": 3148
    }
  ],
  "logging_steps": 500,
  "max_steps": 3148,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 7.801933162066993e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}