checkpoint-500/trainer_state.json

{
  "best_metric": 0.42528408765792847,
  "best_model_checkpoint": "mikhail-panzo/zlm-fil-ceb_b64_le5_s8000/checkpoint-500",
  "epoch": 19.801980198019802,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.9801980198019802,
      "grad_norm": 1.2683665752410889,
      "learning_rate": 2.5000000000000004e-07,
      "loss": 0.4932,
      "step": 50
    },
    {
      "epoch": 3.9603960396039604,
      "grad_norm": 1.1266463994979858,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.4898,
      "step": 100
    },
    {
      "epoch": 5.9405940594059405,
      "grad_norm": 1.1800968647003174,
      "learning_rate": 7.5e-07,
      "loss": 0.4839,
      "step": 150
    },
    {
      "epoch": 7.920792079207921,
      "grad_norm": 0.8235568404197693,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.4785,
      "step": 200
    },
    {
      "epoch": 9.900990099009901,
      "grad_norm": 1.313211441040039,
      "learning_rate": 1.25e-06,
      "loss": 0.4767,
      "step": 250
    },
    {
      "epoch": 11.881188118811881,
      "grad_norm": 0.7831560373306274,
      "learning_rate": 1.5e-06,
      "loss": 0.4681,
      "step": 300
    },
    {
      "epoch": 13.861386138613861,
      "grad_norm": 0.7987237572669983,
      "learning_rate": 1.75e-06,
      "loss": 0.4658,
      "step": 350
    },
    {
      "epoch": 15.841584158415841,
      "grad_norm": 0.7143813371658325,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.4627,
      "step": 400
    },
    {
      "epoch": 17.821782178217823,
      "grad_norm": 0.8037531971931458,
      "learning_rate": 2.25e-06,
      "loss": 0.461,
      "step": 450
    },
    {
      "epoch": 19.801980198019802,
      "grad_norm": 0.79031902551651,
      "learning_rate": 2.5e-06,
      "loss": 0.4592,
      "step": 500
    },
    {
      "epoch": 19.801980198019802,
      "eval_loss": 0.42528408765792847,
      "eval_runtime": 7.8982,
      "eval_samples_per_second": 22.79,
      "eval_steps_per_second": 2.912,
      "step": 500
    }
  ],
  "logging_steps": 50,
  "max_steps": 8000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 320,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5396569205689728.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}