trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 221,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "grad_norm": 14.758803367614746,
      "learning_rate": 0.0002,
      "loss": 23.3472,
      "step": 10
    },
    {
      "epoch": 0.18,
      "grad_norm": 59.627952575683594,
      "learning_rate": 0.0002,
      "loss": 15.2238,
      "step": 20
    },
    {
      "epoch": 0.27,
      "grad_norm": 4.7137131690979,
      "learning_rate": 0.0002,
      "loss": 2.488,
      "step": 30
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.8212027549743652,
      "learning_rate": 0.0002,
      "loss": 1.0811,
      "step": 40
    },
    {
      "epoch": 0.45,
      "grad_norm": 2.546755313873291,
      "learning_rate": 0.0002,
      "loss": 0.8055,
      "step": 50
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.0168941020965576,
      "learning_rate": 0.0002,
      "loss": 0.705,
      "step": 60
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.9273145794868469,
      "learning_rate": 0.0002,
      "loss": 0.6457,
      "step": 70
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.7880547642707825,
      "learning_rate": 0.0002,
      "loss": 0.5931,
      "step": 80
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.8284961581230164,
      "learning_rate": 0.0002,
      "loss": 0.5887,
      "step": 90
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.7326173782348633,
      "learning_rate": 0.0002,
      "loss": 0.5627,
      "step": 100
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.3316080570220947,
      "learning_rate": 0.0002,
      "loss": 0.5699,
      "step": 110
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.7548896074295044,
      "learning_rate": 0.0002,
      "loss": 0.5371,
      "step": 120
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.612474262714386,
      "learning_rate": 0.0002,
      "loss": 0.479,
      "step": 130
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.8262041807174683,
      "learning_rate": 0.0002,
      "loss": 0.5065,
      "step": 140
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.9287819862365723,
      "learning_rate": 0.0002,
      "loss": 0.4441,
      "step": 150
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.7204004526138306,
      "learning_rate": 0.0002,
      "loss": 0.5144,
      "step": 160
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.5320543646812439,
      "learning_rate": 0.0002,
      "loss": 0.4624,
      "step": 170
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.5815476775169373,
      "learning_rate": 0.0002,
      "loss": 0.4536,
      "step": 180
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.4539355933666229,
      "learning_rate": 0.0002,
      "loss": 0.5108,
      "step": 190
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.8424472212791443,
      "learning_rate": 0.0002,
      "loss": 0.4603,
      "step": 200
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.485895574092865,
      "learning_rate": 0.0002,
      "loss": 0.4843,
      "step": 210
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.4039005935192108,
      "learning_rate": 0.0002,
      "loss": 0.4872,
      "step": 220
    }
  ],
  "logging_steps": 10,
  "max_steps": 330,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 6.217168915257754e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}