{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 104,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.40816326530612246,
      "grad_norm": 0.6019407510757446,
      "learning_rate": 8e-05,
      "loss": 3.0962862014770507,
      "step": 5
    },
    {
      "epoch": 0.8163265306122449,
      "grad_norm": 0.418940007686615,
      "learning_rate": 0.00018,
      "loss": 2.6706636428833006,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.1862173080444336,
      "eval_runtime": 3.8648,
      "eval_samples_per_second": 2.07,
      "eval_steps_per_second": 1.035,
      "step": 13
    },
    {
      "epoch": 1.163265306122449,
      "grad_norm": 0.47551143169403076,
      "learning_rate": 0.000199107748815478,
      "loss": 2.476544952392578,
      "step": 15
    },
    {
      "epoch": 1.5714285714285714,
      "grad_norm": 0.41690003871917725,
      "learning_rate": 0.00019551024972069126,
      "loss": 2.1302560806274413,
      "step": 20
    },
    {
      "epoch": 1.9795918367346939,
      "grad_norm": 0.3537037670612335,
      "learning_rate": 0.00018925188358598813,
      "loss": 2.069393539428711,
      "step": 25
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.7015095949172974,
      "eval_runtime": 3.6904,
      "eval_samples_per_second": 2.168,
      "eval_steps_per_second": 1.084,
      "step": 26
    },
    {
      "epoch": 2.326530612244898,
      "grad_norm": 0.3800542950630188,
      "learning_rate": 0.0001805070053127563,
      "loss": 1.9480716705322265,
      "step": 30
    },
    {
      "epoch": 2.7346938775510203,
      "grad_norm": 0.4563523828983307,
      "learning_rate": 0.00016951924276746425,
      "loss": 1.9646163940429688,
      "step": 35
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.6568970680236816,
      "eval_runtime": 3.6916,
      "eval_samples_per_second": 2.167,
      "eval_steps_per_second": 1.084,
      "step": 39
    },
    {
      "epoch": 3.0816326530612246,
      "grad_norm": 0.619397759437561,
      "learning_rate": 0.00015659470943305955,
      "loss": 1.8461959838867188,
      "step": 40
    },
    {
      "epoch": 3.489795918367347,
      "grad_norm": 0.6532426476478577,
      "learning_rate": 0.0001420934762428335,
      "loss": 1.860713005065918,
      "step": 45
    },
    {
      "epoch": 3.8979591836734695,
      "grad_norm": 0.49629712104797363,
      "learning_rate": 0.00012641954018712863,
      "loss": 1.6945793151855468,
      "step": 50
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.645078420639038,
      "eval_runtime": 3.6851,
      "eval_samples_per_second": 2.171,
      "eval_steps_per_second": 1.085,
      "step": 52
    },
    {
      "epoch": 4.244897959183674,
      "grad_norm": 0.5064799785614014,
      "learning_rate": 0.00011000956916240985,
      "loss": 1.9878658294677733,
      "step": 55
    },
    {
      "epoch": 4.653061224489796,
      "grad_norm": 0.633876621723175,
      "learning_rate": 9.332073662548784e-05,
      "loss": 1.9230049133300782,
      "step": 60
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.697495698928833,
      "learning_rate": 7.681798497324716e-05,
      "loss": 1.588405227661133,
      "step": 65
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.6427057981491089,
      "eval_runtime": 3.682,
      "eval_samples_per_second": 2.173,
      "eval_steps_per_second": 1.086,
      "step": 65
    },
    {
      "epoch": 5.408163265306122,
      "grad_norm": 0.6324828267097473,
      "learning_rate": 6.0961072483650526e-05,
      "loss": 1.7133100509643555,
      "step": 70
    },
    {
      "epoch": 5.816326530612245,
      "grad_norm": 0.6433042883872986,
      "learning_rate": 4.6191764683662744e-05,
      "loss": 1.7036520004272462,
      "step": 75
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.6459481716156006,
      "eval_runtime": 3.6793,
      "eval_samples_per_second": 2.174,
      "eval_steps_per_second": 1.087,
      "step": 78
    },
    {
      "epoch": 6.163265306122449,
      "grad_norm": 0.6238939762115479,
      "learning_rate": 3.292152698607768e-05,
      "loss": 1.6959188461303711,
      "step": 80
    },
    {
      "epoch": 6.571428571428571,
      "grad_norm": 0.6813488602638245,
      "learning_rate": 2.1520061472133902e-05,
      "loss": 1.6420322418212892,
      "step": 85
    },
    {
      "epoch": 6.979591836734694,
      "grad_norm": 0.8164582252502441,
      "learning_rate": 1.230500717933285e-05,
      "loss": 1.5098464965820313,
      "step": 90
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.6437904834747314,
      "eval_runtime": 3.6733,
      "eval_samples_per_second": 2.178,
      "eval_steps_per_second": 1.089,
      "step": 91
    },
    {
      "epoch": 7.326530612244898,
      "grad_norm": 0.6925736665725708,
      "learning_rate": 5.533090839208133e-06,
      "loss": 1.703462791442871,
      "step": 95
    },
    {
      "epoch": 7.73469387755102,
      "grad_norm": 0.7122836709022522,
      "learning_rate": 1.3929746009971433e-06,
      "loss": 1.7087106704711914,
      "step": 100
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.6444940567016602,
      "eval_runtime": 3.6947,
      "eval_samples_per_second": 2.165,
      "eval_steps_per_second": 1.083,
      "step": 104
    },
    {
      "epoch": 8.0,
      "step": 104,
      "total_flos": 5533905160949760.0,
      "train_loss": 1.9239307275185218,
      "train_runtime": 911.4517,
      "train_samples_per_second": 0.851,
      "train_steps_per_second": 0.114
    }
  ],
  "logging_steps": 5,
  "max_steps": 104,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5533905160949760.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}