last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.035945363048166784,
  "eval_steps": 4,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0014378145219266715,
      "grad_norm": NaN,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0014378145219266715,
      "eval_loss": NaN,
      "eval_runtime": 10.4334,
      "eval_samples_per_second": 28.083,
      "eval_steps_per_second": 7.093,
      "step": 1
    },
    {
      "epoch": 0.002875629043853343,
      "grad_norm": NaN,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.004313443565780014,
      "grad_norm": NaN,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.005751258087706686,
      "grad_norm": NaN,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.005751258087706686,
      "eval_loss": NaN,
      "eval_runtime": 9.3026,
      "eval_samples_per_second": 31.497,
      "eval_steps_per_second": 7.955,
      "step": 4
    },
    {
      "epoch": 0.007189072609633357,
      "grad_norm": NaN,
      "learning_rate": 5e-06,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.008626887131560028,
      "grad_norm": NaN,
      "learning_rate": 6e-06,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.010064701653486701,
      "grad_norm": NaN,
      "learning_rate": 7e-06,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.011502516175413372,
      "grad_norm": NaN,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.011502516175413372,
      "eval_loss": NaN,
      "eval_runtime": 9.2812,
      "eval_samples_per_second": 31.569,
      "eval_steps_per_second": 7.973,
      "step": 8
    },
    {
      "epoch": 0.012940330697340043,
      "grad_norm": NaN,
      "learning_rate": 9e-06,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.014378145219266714,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.015815959741193385,
      "grad_norm": NaN,
      "learning_rate": 9.890738003669029e-06,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.017253774263120056,
      "grad_norm": NaN,
      "learning_rate": 9.567727288213005e-06,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.017253774263120056,
      "eval_loss": NaN,
      "eval_runtime": 9.288,
      "eval_samples_per_second": 31.546,
      "eval_steps_per_second": 7.967,
      "step": 12
    },
    {
      "epoch": 0.018691588785046728,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-06,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.020129403306973402,
      "grad_norm": NaN,
      "learning_rate": 8.345653031794292e-06,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.021567217828900073,
      "grad_norm": NaN,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.023005032350826744,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-06,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.023005032350826744,
      "eval_loss": NaN,
      "eval_runtime": 9.2594,
      "eval_samples_per_second": 31.644,
      "eval_steps_per_second": 7.992,
      "step": 16
    },
    {
      "epoch": 0.024442846872753415,
      "grad_norm": NaN,
      "learning_rate": 5.522642316338268e-06,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.025880661394680086,
      "grad_norm": NaN,
      "learning_rate": 4.477357683661734e-06,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.027318475916606758,
      "grad_norm": NaN,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.02875629043853343,
      "grad_norm": NaN,
      "learning_rate": 2.5000000000000015e-06,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.02875629043853343,
      "eval_loss": NaN,
      "eval_runtime": 9.2894,
      "eval_samples_per_second": 31.541,
      "eval_steps_per_second": 7.966,
      "step": 20
    },
    {
      "epoch": 0.0301941049604601,
      "grad_norm": NaN,
      "learning_rate": 1.6543469682057105e-06,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.03163191948238677,
      "grad_norm": NaN,
      "learning_rate": 9.549150281252633e-07,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.03306973400431344,
      "grad_norm": NaN,
      "learning_rate": 4.322727117869951e-07,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.03450754852624011,
      "grad_norm": NaN,
      "learning_rate": 1.0926199633097156e-07,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.03450754852624011,
      "eval_loss": NaN,
      "eval_runtime": 9.2745,
      "eval_samples_per_second": 31.592,
      "eval_steps_per_second": 7.979,
      "step": 24
    },
    {
      "epoch": 0.035945363048166784,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 25
    }
  ],
  "logging_steps": 1,
  "max_steps": 25,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1286710965043200.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}