trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9991993594875901,
  "eval_steps": 4,
  "global_step": 78,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.051240992794235385,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.1101,
      "step": 4
    },
    {
      "epoch": 0.10248198558847077,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.085,
      "step": 8
    },
    {
      "epoch": 0.15372297838270615,
      "grad_norm": 0.19277238845825195,
      "learning_rate": 0.002,
      "loss": 1.1055,
      "step": 12
    },
    {
      "epoch": 0.20496397117694154,
      "grad_norm": 0.2435273677110672,
      "learning_rate": 0.0011547005383792518,
      "loss": 1.0199,
      "step": 16
    },
    {
      "epoch": 0.2562049639711769,
      "grad_norm": 1.447454810142517,
      "learning_rate": 0.0008944271909999159,
      "loss": 1.1178,
      "step": 20
    },
    {
      "epoch": 0.3074459567654123,
      "grad_norm": 0.34243044257164,
      "learning_rate": 0.0006666666666666666,
      "loss": 0.8637,
      "step": 24
    },
    {
      "epoch": 0.3586869495596477,
      "grad_norm": 0.18111398816108704,
      "learning_rate": 0.0005547001962252292,
      "loss": 0.8076,
      "step": 28
    },
    {
      "epoch": 0.4099279423538831,
      "grad_norm": 0.16578800976276398,
      "learning_rate": 0.0004850712500726659,
      "loss": 0.7863,
      "step": 32
    },
    {
      "epoch": 0.4611689351481185,
      "grad_norm": 0.18639568984508514,
      "learning_rate": 0.0004364357804719848,
      "loss": 0.7413,
      "step": 36
    },
    {
      "epoch": 0.5124099279423538,
      "grad_norm": 0.29591241478919983,
      "learning_rate": 0.0004,
      "loss": 0.7666,
      "step": 40
    },
    {
      "epoch": 0.5636509207365893,
      "grad_norm": 0.20059671998023987,
      "learning_rate": 0.0003713906763541037,
      "loss": 0.8008,
      "step": 44
    },
    {
      "epoch": 0.6148919135308246,
      "grad_norm": 0.16686993837356567,
      "learning_rate": 0.0003481553119113957,
      "loss": 0.7543,
      "step": 48
    },
    {
      "epoch": 0.6661329063250601,
      "grad_norm": 0.16397202014923096,
      "learning_rate": 0.0003287979746107146,
      "loss": 0.7338,
      "step": 52
    },
    {
      "epoch": 0.7173738991192954,
      "grad_norm": 0.1775909960269928,
      "learning_rate": 0.0003123475237772121,
      "loss": 0.7323,
      "step": 56
    },
    {
      "epoch": 0.7686148919135308,
      "grad_norm": 0.17287716269493103,
      "learning_rate": 0.00029814239699997195,
      "loss": 0.7414,
      "step": 60
    },
    {
      "epoch": 0.8198558847077662,
      "grad_norm": 0.19447582960128784,
      "learning_rate": 0.0002857142857142857,
      "loss": 0.7489,
      "step": 64
    },
    {
      "epoch": 0.8710968775020016,
      "grad_norm": 0.15788140892982483,
      "learning_rate": 0.0002747211278973781,
      "loss": 0.7467,
      "step": 68
    },
    {
      "epoch": 0.922337870296237,
      "grad_norm": 0.18027225136756897,
      "learning_rate": 0.00026490647141300875,
      "loss": 0.734,
      "step": 72
    },
    {
      "epoch": 0.9735788630904724,
      "grad_norm": 0.1561153680086136,
      "learning_rate": 0.000256073759865792,
      "loss": 0.7291,
      "step": 76
    },
    {
      "epoch": 0.9991993594875901,
      "step": 78,
      "total_flos": 7.922041751265608e+17,
      "train_loss": 0.8468386912957216,
      "train_runtime": 778.0462,
      "train_samples_per_second": 12.84,
      "train_steps_per_second": 0.1
    }
  ],
  "logging_steps": 4,
  "max_steps": 78,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 4,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.922041751265608e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}