checkpoint-3000/trainer_state.json

{
  "best_metric": 1.4342260360717773,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l18-l/checkpoint-3000",
  "epoch": 0.3199658703071672,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05332764505119454,
      "grad_norm": 186.70851135253906,
      "learning_rate": 9.866680887372013e-09,
      "loss": 0.4804,
      "step": 500
    },
    {
      "epoch": 0.05332764505119454,
      "eval_loss": 1.4967491626739502,
      "eval_runtime": 60.9128,
      "eval_samples_per_second": 16.204,
      "eval_steps_per_second": 2.036,
      "step": 500
    },
    {
      "epoch": 0.10665529010238908,
      "grad_norm": 86.50623321533203,
      "learning_rate": 9.733361774744027e-09,
      "loss": 0.4928,
      "step": 1000
    },
    {
      "epoch": 0.10665529010238908,
      "eval_loss": 1.482089877128601,
      "eval_runtime": 62.7665,
      "eval_samples_per_second": 15.725,
      "eval_steps_per_second": 1.976,
      "step": 1000
    },
    {
      "epoch": 0.1599829351535836,
      "grad_norm": 603.98974609375,
      "learning_rate": 9.600042662116041e-09,
      "loss": 0.4325,
      "step": 1500
    },
    {
      "epoch": 0.1599829351535836,
      "eval_loss": 1.4686362743377686,
      "eval_runtime": 61.6915,
      "eval_samples_per_second": 15.999,
      "eval_steps_per_second": 2.01,
      "step": 1500
    },
    {
      "epoch": 0.21331058020477817,
      "grad_norm": 13.387365341186523,
      "learning_rate": 9.466723549488055e-09,
      "loss": 0.3964,
      "step": 2000
    },
    {
      "epoch": 0.21331058020477817,
      "eval_loss": 1.4561632871627808,
      "eval_runtime": 61.8331,
      "eval_samples_per_second": 15.962,
      "eval_steps_per_second": 2.005,
      "step": 2000
    },
    {
      "epoch": 0.2666382252559727,
      "grad_norm": 322.0577392578125,
      "learning_rate": 9.333404436860068e-09,
      "loss": 0.4294,
      "step": 2500
    },
    {
      "epoch": 0.2666382252559727,
      "eval_loss": 1.445016622543335,
      "eval_runtime": 63.1527,
      "eval_samples_per_second": 15.629,
      "eval_steps_per_second": 1.963,
      "step": 2500
    },
    {
      "epoch": 0.3199658703071672,
      "grad_norm": 0.010774992406368256,
      "learning_rate": 9.200085324232083e-09,
      "loss": 0.3954,
      "step": 3000
    },
    {
      "epoch": 0.3199658703071672,
      "eval_loss": 1.4342260360717773,
      "eval_runtime": 63.3562,
      "eval_samples_per_second": 15.579,
      "eval_steps_per_second": 1.957,
      "step": 3000
    }
  ],
  "logging_steps": 500,
  "max_steps": 37504,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1079994570482520.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}