last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0023797217956488276,
  "eval_steps": 2000,
  "global_step": 3200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 8.451894760131836,
      "learning_rate": 1.9999999959757473e-05,
      "loss": 1.835,
      "step": 200
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.7373883724212646,
      "learning_rate": 1.9999999832252933e-05,
      "loss": 1.6278,
      "step": 400
    },
    {
      "epoch": 0.0,
      "grad_norm": 3.7490854263305664,
      "learning_rate": 1.9999999617416517e-05,
      "loss": 1.6314,
      "step": 600
    },
    {
      "epoch": 0.0,
      "grad_norm": 10.143038749694824,
      "learning_rate": 1.999999931524823e-05,
      "loss": 1.5416,
      "step": 800
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.783194065093994,
      "learning_rate": 1.999999892574807e-05,
      "loss": 1.5775,
      "step": 1000
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.1446919441223145,
      "learning_rate": 1.9999998448916044e-05,
      "loss": 1.6922,
      "step": 1200
    },
    {
      "epoch": 0.0,
      "grad_norm": 3.6168997287750244,
      "learning_rate": 1.9999997884752155e-05,
      "loss": 1.6211,
      "step": 1400
    },
    {
      "epoch": 0.0,
      "grad_norm": 4.068266868591309,
      "learning_rate": 1.9999997233256404e-05,
      "loss": 1.6001,
      "step": 1600
    },
    {
      "epoch": 0.0,
      "grad_norm": 3.046320676803589,
      "learning_rate": 1.9999996494428805e-05,
      "loss": 1.5682,
      "step": 1800
    },
    {
      "epoch": 0.0,
      "grad_norm": 4.574249267578125,
      "learning_rate": 1.9999995668269356e-05,
      "loss": 1.5658,
      "step": 2000
    },
    {
      "epoch": 0.0,
      "grad_norm": 4.401742935180664,
      "learning_rate": 1.999999475956276e-05,
      "loss": 1.6152,
      "step": 2200
    },
    {
      "epoch": 0.0,
      "grad_norm": 4.141517162322998,
      "learning_rate": 1.9999993759176304e-05,
      "loss": 1.564,
      "step": 2400
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.8213422298431396,
      "learning_rate": 1.9999992671458023e-05,
      "loss": 1.5586,
      "step": 2600
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.3063032627105713,
      "learning_rate": 1.999999149640793e-05,
      "loss": 1.6118,
      "step": 2800
    },
    {
      "epoch": 0.0,
      "grad_norm": 3.5887880325317383,
      "learning_rate": 1.9999990234026036e-05,
      "loss": 1.586,
      "step": 3000
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.8140385150909424,
      "learning_rate": 1.9999988884312347e-05,
      "loss": 1.6221,
      "step": 3200
    }
  ],
  "logging_steps": 200,
  "max_steps": 6723475,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 200,
  "total_flos": 4.180089275793408e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}