{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.978102189781022,
  "eval_steps": 10,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.014598540145985401,
      "eval_loss": 0.3622247874736786,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.4414,
      "eval_samples_per_second": 6.947,
      "eval_steps_per_second": 1.749,
      "step": 1
    },
    {
      "epoch": 0.145985401459854,
      "grad_norm": 50.93666458129883,
      "learning_rate": 5e-06,
      "loss": 1.6392,
      "step": 10
    },
    {
      "epoch": 0.145985401459854,
      "eval_loss": 0.3429737687110901,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.7357,
      "eval_samples_per_second": 6.896,
      "eval_steps_per_second": 1.736,
      "step": 10
    },
    {
      "epoch": 0.291970802919708,
      "grad_norm": 44.70755386352539,
      "learning_rate": 1e-05,
      "loss": 1.1999,
      "step": 20
    },
    {
      "epoch": 0.291970802919708,
      "eval_loss": 0.304928719997406,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.7745,
      "eval_samples_per_second": 6.889,
      "eval_steps_per_second": 1.735,
      "step": 20
    },
    {
      "epoch": 0.43795620437956206,
      "grad_norm": 16.688669204711914,
      "learning_rate": 1.5e-05,
      "loss": 1.2826,
      "step": 30
    },
    {
      "epoch": 0.43795620437956206,
      "eval_loss": 0.2768114507198334,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.9306,
      "eval_samples_per_second": 6.862,
      "eval_steps_per_second": 1.728,
      "step": 30
    },
    {
      "epoch": 0.583941605839416,
      "grad_norm": 25.983545303344727,
      "learning_rate": 2e-05,
      "loss": 0.7583,
      "step": 40
    },
    {
      "epoch": 0.583941605839416,
      "eval_loss": 0.24917608499526978,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.8008,
      "eval_samples_per_second": 6.884,
      "eval_steps_per_second": 1.734,
      "step": 40
    },
    {
      "epoch": 0.7299270072992701,
      "grad_norm": 33.70819091796875,
      "learning_rate": 2.5e-05,
      "loss": 0.603,
      "step": 50
    },
    {
      "epoch": 0.7299270072992701,
      "eval_loss": 0.2258419543504715,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.8305,
      "eval_samples_per_second": 6.879,
      "eval_steps_per_second": 1.732,
      "step": 50
    },
    {
      "epoch": 0.8759124087591241,
      "grad_norm": 22.157100677490234,
      "learning_rate": 3e-05,
      "loss": 1.014,
      "step": 60
    },
    {
      "epoch": 0.8759124087591241,
      "eval_loss": 0.19583500921726227,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.774,
      "eval_samples_per_second": 6.889,
      "eval_steps_per_second": 1.735,
      "step": 60
    },
    {
      "epoch": 1.0218978102189782,
      "grad_norm": 9.168417930603027,
      "learning_rate": 3.5e-05,
      "loss": 0.8131,
      "step": 70
    },
    {
      "epoch": 1.0218978102189782,
      "eval_loss": 0.16876357793807983,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 40.0137,
      "eval_samples_per_second": 6.848,
      "eval_steps_per_second": 1.724,
      "step": 70
    },
    {
      "epoch": 1.167883211678832,
      "grad_norm": 17.78595542907715,
      "learning_rate": 4e-05,
      "loss": 0.6346,
      "step": 80
    },
    {
      "epoch": 1.167883211678832,
      "eval_loss": 0.1590624302625656,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 40.2219,
      "eval_samples_per_second": 6.812,
      "eval_steps_per_second": 1.715,
      "step": 80
    },
    {
      "epoch": 1.313868613138686,
      "grad_norm": 12.886183738708496,
      "learning_rate": 4.5e-05,
      "loss": 0.5089,
      "step": 90
    },
    {
      "epoch": 1.313868613138686,
      "eval_loss": 0.15023882687091827,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 40.1557,
      "eval_samples_per_second": 6.823,
      "eval_steps_per_second": 1.718,
      "step": 90
    },
    {
      "epoch": 1.4598540145985401,
      "grad_norm": 5.830733299255371,
      "learning_rate": 5e-05,
      "loss": 0.4616,
      "step": 100
    },
    {
      "epoch": 1.4598540145985401,
      "eval_loss": 0.1340549886226654,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 40.1143,
      "eval_samples_per_second": 6.83,
      "eval_steps_per_second": 1.72,
      "step": 100
    },
    {
      "epoch": 1.6058394160583942,
      "grad_norm": 8.617661476135254,
      "learning_rate": 4.519230769230769e-05,
      "loss": 0.4498,
      "step": 110
    },
    {
      "epoch": 1.6058394160583942,
      "eval_loss": 0.11359694600105286,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.9036,
      "eval_samples_per_second": 6.867,
      "eval_steps_per_second": 1.729,
      "step": 110
    },
    {
      "epoch": 1.7518248175182483,
      "grad_norm": 120.68909454345703,
      "learning_rate": 4.038461538461539e-05,
      "loss": 0.4422,
      "step": 120
    },
    {
      "epoch": 1.7518248175182483,
      "eval_loss": 0.10619519650936127,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.9903,
      "eval_samples_per_second": 6.852,
      "eval_steps_per_second": 1.725,
      "step": 120
    },
    {
      "epoch": 1.897810218978102,
      "grad_norm": 5.805586338043213,
      "learning_rate": 3.557692307692308e-05,
      "loss": 0.3519,
      "step": 130
    },
    {
      "epoch": 1.897810218978102,
      "eval_loss": 0.09886428713798523,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.9675,
      "eval_samples_per_second": 6.856,
      "eval_steps_per_second": 1.726,
      "step": 130
    },
    {
      "epoch": 2.0437956204379564,
      "grad_norm": 11.567919731140137,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.2382,
      "step": 140
    },
    {
      "epoch": 2.0437956204379564,
      "eval_loss": 0.09246355295181274,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.9889,
      "eval_samples_per_second": 6.852,
      "eval_steps_per_second": 1.725,
      "step": 140
    },
    {
      "epoch": 2.18978102189781,
      "grad_norm": 38.91920852661133,
      "learning_rate": 2.5961538461538464e-05,
      "loss": 0.242,
      "step": 150
    },
    {
      "epoch": 2.18978102189781,
      "eval_loss": 0.08943724632263184,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 40.1851,
      "eval_samples_per_second": 6.818,
      "eval_steps_per_second": 1.717,
      "step": 150
    },
    {
      "epoch": 2.335766423357664,
      "grad_norm": 0.6526745557785034,
      "learning_rate": 2.1153846153846154e-05,
      "loss": 0.3462,
      "step": 160
    },
    {
      "epoch": 2.335766423357664,
      "eval_loss": 0.09070777893066406,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 40.0427,
      "eval_samples_per_second": 6.843,
      "eval_steps_per_second": 1.723,
      "step": 160
    },
    {
      "epoch": 2.4817518248175183,
      "grad_norm": 9.621135711669922,
      "learning_rate": 1.6346153846153847e-05,
      "loss": 0.1371,
      "step": 170
    },
    {
      "epoch": 2.4817518248175183,
      "eval_loss": 0.08618415892124176,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.9436,
      "eval_samples_per_second": 6.86,
      "eval_steps_per_second": 1.727,
      "step": 170
    },
    {
      "epoch": 2.627737226277372,
      "grad_norm": 20.143051147460938,
      "learning_rate": 1.153846153846154e-05,
      "loss": 0.2691,
      "step": 180
    },
    {
      "epoch": 2.627737226277372,
      "eval_loss": 0.08378317207098007,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 40.04,
      "eval_samples_per_second": 6.843,
      "eval_steps_per_second": 1.723,
      "step": 180
    },
    {
      "epoch": 2.7737226277372264,
      "grad_norm": 0.41237956285476685,
      "learning_rate": 6.730769230769231e-06,
      "loss": 0.0869,
      "step": 190
    },
    {
      "epoch": 2.7737226277372264,
      "eval_loss": 0.08326868712902069,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.9047,
      "eval_samples_per_second": 6.866,
      "eval_steps_per_second": 1.729,
      "step": 190
    },
    {
      "epoch": 2.9197080291970803,
      "grad_norm": 0.11198130249977112,
      "learning_rate": 1.9230769230769234e-06,
      "loss": 0.3401,
      "step": 200
    },
    {
      "epoch": 2.9197080291970803,
      "eval_loss": 0.0814913883805275,
      "eval_model_preparation_time": 0.0061,
      "eval_runtime": 39.93,
      "eval_samples_per_second": 6.862,
      "eval_steps_per_second": 1.728,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 204,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9136781869568e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}