trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8417508417508418,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04208754208754209,
      "grad_norm": 44.5,
      "learning_rate": 2.222222222222222e-06,
      "loss": 9.708373413085937,
      "step": 25
    },
    {
      "epoch": 0.08417508417508418,
      "grad_norm": 20.5,
      "learning_rate": 4.537037037037038e-06,
      "loss": 7.022988891601562,
      "step": 50
    },
    {
      "epoch": 0.12626262626262627,
      "grad_norm": 13.375,
      "learning_rate": 4.94212962962963e-06,
      "loss": 4.4798516845703125,
      "step": 75
    },
    {
      "epoch": 0.16835016835016836,
      "grad_norm": 35.75,
      "learning_rate": 4.869791666666667e-06,
      "loss": 3.579981384277344,
      "step": 100
    },
    {
      "epoch": 0.21043771043771045,
      "grad_norm": 33.0,
      "learning_rate": 4.797453703703704e-06,
      "loss": 2.9001919555664064,
      "step": 125
    },
    {
      "epoch": 0.25252525252525254,
      "grad_norm": 38.5,
      "learning_rate": 4.725115740740741e-06,
      "loss": 2.722982177734375,
      "step": 150
    },
    {
      "epoch": 0.2946127946127946,
      "grad_norm": 98.0,
      "learning_rate": 4.652777777777779e-06,
      "loss": 2.658799133300781,
      "step": 175
    },
    {
      "epoch": 0.3367003367003367,
      "grad_norm": 37.5,
      "learning_rate": 4.580439814814815e-06,
      "loss": 2.5711898803710938,
      "step": 200
    },
    {
      "epoch": 0.3787878787878788,
      "grad_norm": 148.0,
      "learning_rate": 4.508101851851852e-06,
      "loss": 2.5608544921875,
      "step": 225
    },
    {
      "epoch": 0.4208754208754209,
      "grad_norm": 100.0,
      "learning_rate": 4.435763888888889e-06,
      "loss": 2.5246621704101564,
      "step": 250
    },
    {
      "epoch": 0.46296296296296297,
      "grad_norm": 111.0,
      "learning_rate": 4.363425925925927e-06,
      "loss": 2.504690399169922,
      "step": 275
    },
    {
      "epoch": 0.5050505050505051,
      "grad_norm": 74.5,
      "learning_rate": 4.2910879629629634e-06,
      "loss": 2.456527099609375,
      "step": 300
    },
    {
      "epoch": 0.5471380471380471,
      "grad_norm": 84.0,
      "learning_rate": 4.21875e-06,
      "loss": 2.476842498779297,
      "step": 325
    },
    {
      "epoch": 0.5892255892255892,
      "grad_norm": 26.125,
      "learning_rate": 4.1464120370370375e-06,
      "loss": 2.447265930175781,
      "step": 350
    },
    {
      "epoch": 0.6313131313131313,
      "grad_norm": 38.5,
      "learning_rate": 4.074074074074074e-06,
      "loss": 2.3707008361816406,
      "step": 375
    },
    {
      "epoch": 0.6734006734006734,
      "grad_norm": 23.375,
      "learning_rate": 4.001736111111112e-06,
      "loss": 2.374063720703125,
      "step": 400
    },
    {
      "epoch": 0.7154882154882155,
      "grad_norm": 21.875,
      "learning_rate": 3.929398148148148e-06,
      "loss": 2.3581019592285157,
      "step": 425
    },
    {
      "epoch": 0.7575757575757576,
      "grad_norm": 21.25,
      "learning_rate": 3.857060185185186e-06,
      "loss": 2.3752615356445315,
      "step": 450
    },
    {
      "epoch": 0.7996632996632996,
      "grad_norm": 48.0,
      "learning_rate": 3.7847222222222224e-06,
      "loss": 2.309081726074219,
      "step": 475
    },
    {
      "epoch": 0.8417508417508418,
      "grad_norm": 17.375,
      "learning_rate": 3.7123842592592594e-06,
      "loss": 2.265882568359375,
      "step": 500
    }
  ],
  "logging_steps": 25,
  "max_steps": 1782,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 4.59351584145408e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}