last-checkpoint/trainer_state.json

{
  "best_metric": 1.1258819103240967,
  "best_model_checkpoint": "miner_id_24/checkpoint-50",
  "epoch": 0.05092514004413512,
  "eval_steps": 50,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00033950093362756747,
      "eval_loss": 1.233475923538208,
      "eval_runtime": 73.3685,
      "eval_samples_per_second": 16.915,
      "eval_steps_per_second": 4.239,
      "step": 1
    },
    {
      "epoch": 0.003395009336275675,
      "grad_norm": 8.238821983337402,
      "learning_rate": 4.34e-05,
      "loss": 1.651,
      "step": 10
    },
    {
      "epoch": 0.00679001867255135,
      "grad_norm": 8.250868797302246,
      "learning_rate": 8.68e-05,
      "loss": 1.7066,
      "step": 20
    },
    {
      "epoch": 0.010185028008827025,
      "grad_norm": 10.320452690124512,
      "learning_rate": 0.0001302,
      "loss": 1.6893,
      "step": 30
    },
    {
      "epoch": 0.0135800373451027,
      "grad_norm": 9.33686351776123,
      "learning_rate": 0.0001736,
      "loss": 1.6924,
      "step": 40
    },
    {
      "epoch": 0.016975046681378374,
      "grad_norm": 21.88421630859375,
      "learning_rate": 0.000217,
      "loss": 1.9382,
      "step": 50
    },
    {
      "epoch": 0.016975046681378374,
      "eval_loss": 1.1258819103240967,
      "eval_runtime": 73.3246,
      "eval_samples_per_second": 16.925,
      "eval_steps_per_second": 4.241,
      "step": 50
    },
    {
      "epoch": 0.02037005601765405,
      "grad_norm": 7.034121513366699,
      "learning_rate": 0.00021673569945319091,
      "loss": 1.7911,
      "step": 60
    },
    {
      "epoch": 0.023765065353929724,
      "grad_norm": 7.012728214263916,
      "learning_rate": 0.00021594408545846038,
      "loss": 2.0025,
      "step": 70
    },
    {
      "epoch": 0.0271600746902054,
      "grad_norm": 6.738110065460205,
      "learning_rate": 0.0002146290146796179,
      "loss": 2.0042,
      "step": 80
    },
    {
      "epoch": 0.030555084026481074,
      "grad_norm": 9.321654319763184,
      "learning_rate": 0.0002127968940093076,
      "loss": 2.0401,
      "step": 90
    },
    {
      "epoch": 0.03395009336275675,
      "grad_norm": 29.11263656616211,
      "learning_rate": 0.00021045664935527106,
      "loss": 2.1941,
      "step": 100
    },
    {
      "epoch": 0.03395009336275675,
      "eval_loss": 1.2958543300628662,
      "eval_runtime": 73.5899,
      "eval_samples_per_second": 16.864,
      "eval_steps_per_second": 4.226,
      "step": 100
    },
    {
      "epoch": 0.03734510269903242,
      "grad_norm": 6.313995838165283,
      "learning_rate": 0.00020761968215422217,
      "loss": 2.0996,
      "step": 110
    },
    {
      "epoch": 0.0407401120353081,
      "grad_norm": 7.566165924072266,
      "learning_rate": 0.00020429981382519356,
      "loss": 2.104,
      "step": 120
    },
    {
      "epoch": 0.04413512137158377,
      "grad_norm": 8.9271240234375,
      "learning_rate": 0.00020051321843297219,
      "loss": 2.0573,
      "step": 130
    },
    {
      "epoch": 0.04753013070785945,
      "grad_norm": 10.48095989227295,
      "learning_rate": 0.0001962783438896818,
      "loss": 2.1817,
      "step": 140
    },
    {
      "epoch": 0.05092514004413512,
      "grad_norm": 31.889493942260742,
      "learning_rate": 0.0001916158220784091,
      "loss": 2.0493,
      "step": 150
    },
    {
      "epoch": 0.05092514004413512,
      "eval_loss": 1.2741807699203491,
      "eval_runtime": 73.4449,
      "eval_samples_per_second": 16.897,
      "eval_steps_per_second": 4.234,
      "step": 150
    }
  ],
  "logging_steps": 10,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 2
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.692065004408013e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}