last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.05711022272986865,
  "eval_steps": 5,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002284408909194746,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.002284408909194746,
      "eval_loss": NaN,
      "eval_runtime": 121.7405,
      "eval_samples_per_second": 3.031,
      "eval_steps_per_second": 1.52,
      "step": 1
    },
    {
      "epoch": 0.004568817818389492,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.006853226727584237,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.009137635636778984,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.011422044545973729,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.011422044545973729,
      "eval_loss": NaN,
      "eval_runtime": 121.9164,
      "eval_samples_per_second": 3.027,
      "eval_steps_per_second": 1.517,
      "step": 5
    },
    {
      "epoch": 0.013706453455168474,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.01599086236436322,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.018275271273557967,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.020559680182752713,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.022844089091947458,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.022844089091947458,
      "eval_loss": NaN,
      "eval_runtime": 122.0054,
      "eval_samples_per_second": 3.024,
      "eval_steps_per_second": 1.516,
      "step": 10
    },
    {
      "epoch": 0.025128498001142203,
      "grad_norm": NaN,
      "learning_rate": 0.0001996917333733128,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.02741290691033695,
      "grad_norm": NaN,
      "learning_rate": 0.00019876883405951377,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.029697315819531698,
      "grad_norm": NaN,
      "learning_rate": 0.00019723699203976766,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.03198172472872644,
      "grad_norm": NaN,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.034266133637921185,
      "grad_norm": NaN,
      "learning_rate": 0.0001923879532511287,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.034266133637921185,
      "eval_loss": NaN,
      "eval_runtime": 122.1464,
      "eval_samples_per_second": 3.021,
      "eval_steps_per_second": 1.515,
      "step": 15
    },
    {
      "epoch": 0.036550542547115934,
      "grad_norm": NaN,
      "learning_rate": 0.0001891006524188368,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.038834951456310676,
      "grad_norm": NaN,
      "learning_rate": 0.00018526401643540922,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.041119360365505425,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.043403769274700174,
      "grad_norm": NaN,
      "learning_rate": 0.0001760405965600031,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.045688178183894916,
      "grad_norm": NaN,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.045688178183894916,
      "eval_loss": NaN,
      "eval_runtime": 122.164,
      "eval_samples_per_second": 3.021,
      "eval_steps_per_second": 1.514,
      "step": 20
    },
    {
      "epoch": 0.047972587093089665,
      "grad_norm": NaN,
      "learning_rate": 0.00016494480483301836,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.05025699600228441,
      "grad_norm": NaN,
      "learning_rate": 0.00015877852522924732,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.052541404911479156,
      "grad_norm": NaN,
      "learning_rate": 0.0001522498564715949,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.0548258138206739,
      "grad_norm": NaN,
      "learning_rate": 0.00014539904997395468,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.05711022272986865,
      "grad_norm": NaN,
      "learning_rate": 0.000138268343236509,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.05711022272986865,
      "eval_loss": NaN,
      "eval_runtime": 122.2721,
      "eval_samples_per_second": 3.018,
      "eval_steps_per_second": 1.513,
      "step": 25
    }
  ],
  "logging_steps": 1,
  "max_steps": 50,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.7182011785216e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}