trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 2244,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13368983957219252,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 100
    },
    {
      "epoch": 0.26737967914438504,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 200
    },
    {
      "epoch": 0.40106951871657753,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 300
    },
    {
      "epoch": 0.5347593582887701,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 400
    },
    {
      "epoch": 0.6684491978609626,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 500
    },
    {
      "epoch": 0.6684491978609626,
      "eval_runtime": 23.2889,
      "eval_samples_per_second": 64.28,
      "eval_steps_per_second": 32.161,
      "step": 500
    },
    {
      "epoch": 0.8021390374331551,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 600
    },
    {
      "epoch": 0.9358288770053476,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 700
    },
    {
      "epoch": 1.0695187165775402,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 800
    },
    {
      "epoch": 1.2032085561497325,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 900
    },
    {
      "epoch": 1.3368983957219251,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 1000
    },
    {
      "epoch": 1.3368983957219251,
      "eval_runtime": 23.0055,
      "eval_samples_per_second": 65.071,
      "eval_steps_per_second": 32.557,
      "step": 1000
    },
    {
      "epoch": 1.4705882352941178,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 1100
    },
    {
      "epoch": 1.6042780748663101,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 1200
    },
    {
      "epoch": 1.7379679144385025,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 1300
    },
    {
      "epoch": 1.8716577540106951,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 1400
    },
    {
      "epoch": 2.0053475935828877,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 1500
    },
    {
      "epoch": 2.0053475935828877,
      "eval_runtime": 22.6355,
      "eval_samples_per_second": 66.135,
      "eval_steps_per_second": 33.09,
      "step": 1500
    },
    {
      "epoch": 2.1390374331550803,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 1600
    },
    {
      "epoch": 2.2727272727272725,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 1700
    },
    {
      "epoch": 2.406417112299465,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 1800
    },
    {
      "epoch": 2.5401069518716577,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 1900
    },
    {
      "epoch": 2.6737967914438503,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 2000
    },
    {
      "epoch": 2.6737967914438503,
      "eval_runtime": 22.6195,
      "eval_samples_per_second": 66.182,
      "eval_steps_per_second": 33.113,
      "step": 2000
    },
    {
      "epoch": 2.807486631016043,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 2100
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 2200
    }
  ],
  "logging_steps": 100,
  "max_steps": 2244,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 680410356297408.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}