last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.008438818565400843,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008438818565400844,
      "grad_norm": 0.31494563817977905,
      "learning_rate": 2e-05,
      "loss": 2.6776,
      "step": 1
    },
    {
      "epoch": 0.0008438818565400844,
      "eval_loss": 0.6653371453285217,
      "eval_runtime": 57.408,
      "eval_samples_per_second": 8.692,
      "eval_steps_per_second": 4.355,
      "step": 1
    },
    {
      "epoch": 0.0016877637130801688,
      "grad_norm": 0.31404587626457214,
      "learning_rate": 4e-05,
      "loss": 2.8598,
      "step": 2
    },
    {
      "epoch": 0.002531645569620253,
      "grad_norm": 0.28022581338882446,
      "learning_rate": 6e-05,
      "loss": 2.4423,
      "step": 3
    },
    {
      "epoch": 0.002531645569620253,
      "eval_loss": 0.6648818850517273,
      "eval_runtime": 57.5253,
      "eval_samples_per_second": 8.674,
      "eval_steps_per_second": 4.346,
      "step": 3
    },
    {
      "epoch": 0.0033755274261603376,
      "grad_norm": 0.36158403754234314,
      "learning_rate": 8e-05,
      "loss": 2.6595,
      "step": 4
    },
    {
      "epoch": 0.004219409282700422,
      "grad_norm": 0.3164409101009369,
      "learning_rate": 0.0001,
      "loss": 2.7197,
      "step": 5
    },
    {
      "epoch": 0.005063291139240506,
      "grad_norm": 0.35253241658210754,
      "learning_rate": 0.00012,
      "loss": 2.6898,
      "step": 6
    },
    {
      "epoch": 0.005063291139240506,
      "eval_loss": 0.656719446182251,
      "eval_runtime": 57.6211,
      "eval_samples_per_second": 8.66,
      "eval_steps_per_second": 4.339,
      "step": 6
    },
    {
      "epoch": 0.00590717299578059,
      "grad_norm": 0.3555045425891876,
      "learning_rate": 0.00014,
      "loss": 2.464,
      "step": 7
    },
    {
      "epoch": 0.006751054852320675,
      "grad_norm": 0.3778963088989258,
      "learning_rate": 0.00016,
      "loss": 2.2624,
      "step": 8
    },
    {
      "epoch": 0.007594936708860759,
      "grad_norm": 0.4649808704853058,
      "learning_rate": 0.00018,
      "loss": 2.4903,
      "step": 9
    },
    {
      "epoch": 0.007594936708860759,
      "eval_loss": 0.6111209988594055,
      "eval_runtime": 57.5504,
      "eval_samples_per_second": 8.671,
      "eval_steps_per_second": 4.344,
      "step": 9
    },
    {
      "epoch": 0.008438818565400843,
      "grad_norm": 0.447596937417984,
      "learning_rate": 0.0002,
      "loss": 2.4649,
      "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": 6794734731264000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}