last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03590664272890485,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003590664272890485,
      "grad_norm": 3.4567840099334717,
      "learning_rate": 2e-05,
      "loss": 43.9531,
      "step": 1
    },
    {
      "epoch": 0.00718132854578097,
      "grad_norm": 3.462780714035034,
      "learning_rate": 4e-05,
      "loss": 43.9955,
      "step": 2
    },
    {
      "epoch": 0.010771992818671455,
      "grad_norm": 3.8066537380218506,
      "learning_rate": 6e-05,
      "loss": 43.9444,
      "step": 3
    },
    {
      "epoch": 0.010771992818671455,
      "eval_loss": 10.990278244018555,
      "eval_runtime": 1.9558,
      "eval_samples_per_second": 60.332,
      "eval_steps_per_second": 30.166,
      "step": 3
    },
    {
      "epoch": 0.01436265709156194,
      "grad_norm": 3.517993211746216,
      "learning_rate": 8e-05,
      "loss": 43.9742,
      "step": 4
    },
    {
      "epoch": 0.017953321364452424,
      "grad_norm": 3.6830387115478516,
      "learning_rate": 0.0001,
      "loss": 43.9514,
      "step": 5
    },
    {
      "epoch": 0.02154398563734291,
      "grad_norm": 3.59987211227417,
      "learning_rate": 0.00012,
      "loss": 43.9096,
      "step": 6
    },
    {
      "epoch": 0.02154398563734291,
      "eval_loss": 10.98271369934082,
      "eval_runtime": 1.3856,
      "eval_samples_per_second": 85.163,
      "eval_steps_per_second": 42.581,
      "step": 6
    },
    {
      "epoch": 0.025134649910233394,
      "grad_norm": 3.461907386779785,
      "learning_rate": 0.00014,
      "loss": 43.9099,
      "step": 7
    },
    {
      "epoch": 0.02872531418312388,
      "grad_norm": 3.5759644508361816,
      "learning_rate": 0.00016,
      "loss": 43.9183,
      "step": 8
    },
    {
      "epoch": 0.03231597845601436,
      "grad_norm": 3.42861008644104,
      "learning_rate": 0.00018,
      "loss": 43.8975,
      "step": 9
    },
    {
      "epoch": 0.03231597845601436,
      "eval_loss": 10.96753215789795,
      "eval_runtime": 1.1907,
      "eval_samples_per_second": 99.105,
      "eval_steps_per_second": 49.553,
      "step": 9
    },
    {
      "epoch": 0.03590664272890485,
      "grad_norm": 3.638361692428589,
      "learning_rate": 0.0002,
      "loss": 43.8576,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7471104000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}