trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 10000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 0.03134724870324135,
      "learning_rate": 0.00095,
      "loss": 1.1746,
      "step": 500
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.01849432662129402,
      "learning_rate": 0.0009000000000000001,
      "loss": 1.1003,
      "step": 1000
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.03626084700226784,
      "learning_rate": 0.00085,
      "loss": 1.0882,
      "step": 1500
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.03540480509400368,
      "learning_rate": 0.0008,
      "loss": 1.0733,
      "step": 2000
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.04738762974739075,
      "learning_rate": 0.00075,
      "loss": 1.0577,
      "step": 2500
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.06647106260061264,
      "learning_rate": 0.0007,
      "loss": 1.059,
      "step": 3000
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.06651030480861664,
      "learning_rate": 0.0006500000000000001,
      "loss": 1.0458,
      "step": 3500
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.07928076386451721,
      "learning_rate": 0.0006,
      "loss": 1.0317,
      "step": 4000
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.09158056229352951,
      "learning_rate": 0.00055,
      "loss": 1.0277,
      "step": 4500
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.08878821134567261,
      "learning_rate": 0.0005,
      "loss": 1.0124,
      "step": 5000
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.13431012630462646,
      "learning_rate": 0.00045000000000000004,
      "loss": 1.0007,
      "step": 5500
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.10832180082798004,
      "learning_rate": 0.0004,
      "loss": 0.9889,
      "step": 6000
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.10065455734729767,
      "learning_rate": 0.00035,
      "loss": 0.9858,
      "step": 6500
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.1227111890912056,
      "learning_rate": 0.0003,
      "loss": 0.971,
      "step": 7000
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.13141539692878723,
      "learning_rate": 0.00025,
      "loss": 0.9733,
      "step": 7500
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.1325940638780594,
      "learning_rate": 0.0002,
      "loss": 0.9567,
      "step": 8000
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.13246145844459534,
      "learning_rate": 0.00015,
      "loss": 0.9713,
      "step": 8500
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.12682273983955383,
      "learning_rate": 0.0001,
      "loss": 0.9513,
      "step": 9000
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.15320372581481934,
      "learning_rate": 5e-05,
      "loss": 0.9449,
      "step": 9500
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.15557877719402313,
      "learning_rate": 0.0,
      "loss": 0.9535,
      "step": 10000
    }
  ],
  "logging_steps": 500,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 2000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.4824605073408e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}