checkpoint_360/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7331975560081466,
  "eval_steps": 500,
  "global_step": 360,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.6532,
      "step": 10
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.4479,
      "step": 20
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.5807,
      "step": 30
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.8864,
      "step": 40
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 1.9486,
      "step": 50
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 1.4537,
      "step": 60
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 1.4056,
      "step": 70
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002,
      "loss": 1.4739,
      "step": 80
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002,
      "loss": 1.878,
      "step": 90
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 1.9504,
      "step": 100
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 1.465,
      "step": 110
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 1.4416,
      "step": 120
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002,
      "loss": 1.5208,
      "step": 130
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 1.8197,
      "step": 140
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 1.9127,
      "step": 150
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002,
      "loss": 1.4908,
      "step": 160
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 1.3915,
      "step": 170
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002,
      "loss": 1.5791,
      "step": 180
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002,
      "loss": 2.0458,
      "step": 190
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002,
      "loss": 1.9205,
      "step": 200
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002,
      "loss": 1.4595,
      "step": 210
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002,
      "loss": 1.4085,
      "step": 220
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002,
      "loss": 1.4964,
      "step": 230
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002,
      "loss": 1.8586,
      "step": 240
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 1.8511,
      "step": 250
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 1.3773,
      "step": 260
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002,
      "loss": 1.3869,
      "step": 270
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002,
      "loss": 1.6083,
      "step": 280
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002,
      "loss": 1.8549,
      "step": 290
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002,
      "loss": 1.8714,
      "step": 300
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 1.3614,
      "step": 310
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002,
      "loss": 1.4,
      "step": 320
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002,
      "loss": 1.5435,
      "step": 330
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002,
      "loss": 1.8155,
      "step": 340
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002,
      "loss": 1.9587,
      "step": 350
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 1.3705,
      "step": 360
    }
  ],
  "logging_steps": 10,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 20,
  "total_flos": 1.757604963096576e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}