trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 184,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21739130434782608,
      "grad_norm": 98.25260162353516,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 3.2276,
      "step": 10
    },
    {
      "epoch": 0.43478260869565216,
      "grad_norm": 77.15660858154297,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 2.4812,
      "step": 20
    },
    {
      "epoch": 0.6521739130434783,
      "grad_norm": 44.03319549560547,
      "learning_rate": 3e-06,
      "loss": 2.4834,
      "step": 30
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 36.34611892700195,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.8211,
      "step": 40
    },
    {
      "epoch": 1.0869565217391304,
      "grad_norm": 26.440757751464844,
      "learning_rate": 5e-06,
      "loss": 1.8561,
      "step": 50
    },
    {
      "epoch": 1.3043478260869565,
      "grad_norm": 31.280738830566406,
      "learning_rate": 6e-06,
      "loss": 1.3881,
      "step": 60
    },
    {
      "epoch": 1.5217391304347827,
      "grad_norm": 18.75351333618164,
      "learning_rate": 7e-06,
      "loss": 1.1658,
      "step": 70
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 31.8607177734375,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.4099,
      "step": 80
    },
    {
      "epoch": 1.9565217391304348,
      "grad_norm": 40.35784912109375,
      "learning_rate": 9e-06,
      "loss": 1.2576,
      "step": 90
    },
    {
      "epoch": 2.1739130434782608,
      "grad_norm": 14.189046859741211,
      "learning_rate": 1e-05,
      "loss": 0.9798,
      "step": 100
    },
    {
      "epoch": 2.391304347826087,
      "grad_norm": 12.539738655090332,
      "learning_rate": 8.80952380952381e-06,
      "loss": 1.0279,
      "step": 110
    },
    {
      "epoch": 2.608695652173913,
      "grad_norm": 11.521944046020508,
      "learning_rate": 7.61904761904762e-06,
      "loss": 0.7695,
      "step": 120
    },
    {
      "epoch": 2.8260869565217392,
      "grad_norm": 16.423460006713867,
      "learning_rate": 6.4285714285714295e-06,
      "loss": 0.8634,
      "step": 130
    },
    {
      "epoch": 3.0434782608695654,
      "grad_norm": 14.004690170288086,
      "learning_rate": 5.2380952380952384e-06,
      "loss": 0.8086,
      "step": 140
    },
    {
      "epoch": 3.260869565217391,
      "grad_norm": 19.613393783569336,
      "learning_rate": 4.047619047619048e-06,
      "loss": 0.6529,
      "step": 150
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 10.613371849060059,
      "learning_rate": 2.8571428571428573e-06,
      "loss": 0.6305,
      "step": 160
    },
    {
      "epoch": 3.6956521739130435,
      "grad_norm": 19.775999069213867,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 0.5669,
      "step": 170
    },
    {
      "epoch": 3.9130434782608696,
      "grad_norm": 22.53849220275879,
      "learning_rate": 4.7619047619047623e-07,
      "loss": 0.6313,
      "step": 180
    }
  ],
  "logging_steps": 10,
  "max_steps": 184,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 285336058355712.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}