checkpoint-200/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.2738853503184713,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06369426751592357,
      "grad_norm": 21.267967714208993,
      "learning_rate": 1.8750000000000003e-06,
      "loss": 2.447,
      "step": 10
    },
    {
      "epoch": 0.12738853503184713,
      "grad_norm": 4.5780419436683655,
      "learning_rate": 3.958333333333333e-06,
      "loss": 1.8303,
      "step": 20
    },
    {
      "epoch": 0.1910828025477707,
      "grad_norm": 2.6941629656600843,
      "learning_rate": 6.041666666666667e-06,
      "loss": 1.2619,
      "step": 30
    },
    {
      "epoch": 0.25477707006369427,
      "grad_norm": 2.212026239851483,
      "learning_rate": 8.125000000000001e-06,
      "loss": 1.0527,
      "step": 40
    },
    {
      "epoch": 0.3184713375796178,
      "grad_norm": 2.201982410023539,
      "learning_rate": 9.999862102299874e-06,
      "loss": 0.958,
      "step": 50
    },
    {
      "epoch": 0.3821656050955414,
      "grad_norm": 1.784738906519362,
      "learning_rate": 9.983323579940351e-06,
      "loss": 0.917,
      "step": 60
    },
    {
      "epoch": 0.445859872611465,
      "grad_norm": 1.8089946959171528,
      "learning_rate": 9.939310009499348e-06,
      "loss": 0.893,
      "step": 70
    },
    {
      "epoch": 0.5095541401273885,
      "grad_norm": 1.9080843392439555,
      "learning_rate": 9.868064055324204e-06,
      "loss": 0.8733,
      "step": 80
    },
    {
      "epoch": 0.5732484076433121,
      "grad_norm": 1.7471544376369283,
      "learning_rate": 9.76997852474223e-06,
      "loss": 0.8497,
      "step": 90
    },
    {
      "epoch": 0.6369426751592356,
      "grad_norm": 1.7890465333644847,
      "learning_rate": 9.645594202357438e-06,
      "loss": 0.8419,
      "step": 100
    },
    {
      "epoch": 0.7006369426751592,
      "grad_norm": 1.6554683912315524,
      "learning_rate": 9.495596868489588e-06,
      "loss": 0.8287,
      "step": 110
    },
    {
      "epoch": 0.7643312101910829,
      "grad_norm": 1.9772255855653391,
      "learning_rate": 9.320813518194084e-06,
      "loss": 0.8228,
      "step": 120
    },
    {
      "epoch": 0.8280254777070064,
      "grad_norm": 1.6826525630629694,
      "learning_rate": 9.122207801708802e-06,
      "loss": 0.7981,
      "step": 130
    },
    {
      "epoch": 0.89171974522293,
      "grad_norm": 1.6662718650026702,
      "learning_rate": 8.900874711466436e-06,
      "loss": 0.7887,
      "step": 140
    },
    {
      "epoch": 0.9554140127388535,
      "grad_norm": 1.8810043934603964,
      "learning_rate": 8.658034544965003e-06,
      "loss": 0.7796,
      "step": 150
    },
    {
      "epoch": 1.019108280254777,
      "grad_norm": 1.9585298536176439,
      "learning_rate": 8.395026176781627e-06,
      "loss": 0.7628,
      "step": 160
    },
    {
      "epoch": 1.0828025477707006,
      "grad_norm": 1.7354782145978755,
      "learning_rate": 8.113299676823614e-06,
      "loss": 0.7154,
      "step": 170
    },
    {
      "epoch": 1.1464968152866242,
      "grad_norm": 1.7084830286293966,
      "learning_rate": 7.814408315515419e-06,
      "loss": 0.7106,
      "step": 180
    },
    {
      "epoch": 1.2101910828025477,
      "grad_norm": 1.825987376263087,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.7161,
      "step": 190
    },
    {
      "epoch": 1.2738853503184713,
      "grad_norm": 1.9190047988376515,
      "learning_rate": 7.1718081885702905e-06,
      "loss": 0.6945,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 471,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 15387555004416.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}