{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0683053040103494,
  "eval_steps": 250,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0517464424320828,
      "grad_norm": 17.125,
      "learning_rate": 3.356164383561644e-06,
      "loss": 3.4162884521484376,
      "step": 50
    },
    {
      "epoch": 0.1034928848641656,
      "grad_norm": 17.25,
      "learning_rate": 6.78082191780822e-06,
      "loss": 1.7713389587402344,
      "step": 100
    },
    {
      "epoch": 0.15523932729624837,
      "grad_norm": 9.8125,
      "learning_rate": 9.999970742404346e-06,
      "loss": 1.523115997314453,
      "step": 150
    },
    {
      "epoch": 0.2069857697283312,
      "grad_norm": 11.1875,
      "learning_rate": 9.990871149623236e-06,
      "loss": 1.4373092651367188,
      "step": 200
    },
    {
      "epoch": 0.258732212160414,
      "grad_norm": 6.9375,
      "learning_rate": 9.965551392499958e-06,
      "loss": 1.3974497985839844,
      "step": 250
    },
    {
      "epoch": 0.258732212160414,
      "eval_loss": 1.3120511770248413,
      "eval_runtime": 185.4563,
      "eval_samples_per_second": 53.921,
      "eval_steps_per_second": 0.75,
      "step": 250
    },
    {
      "epoch": 0.31047865459249674,
      "grad_norm": 6.90625,
      "learning_rate": 9.924093759398476e-06,
      "loss": 1.3703665161132812,
      "step": 300
    },
    {
      "epoch": 0.36222509702457956,
      "grad_norm": 8.125,
      "learning_rate": 9.86663298624003e-06,
      "loss": 1.35594482421875,
      "step": 350
    },
    {
      "epoch": 0.4139715394566624,
      "grad_norm": 6.75,
      "learning_rate": 9.793355818615905e-06,
      "loss": 1.3408810424804687,
      "step": 400
    },
    {
      "epoch": 0.46571798188874514,
      "grad_norm": 7.75,
      "learning_rate": 9.7045004048705e-06,
      "loss": 1.3330262756347657,
      "step": 450
    },
    {
      "epoch": 0.517464424320828,
      "grad_norm": 10.0,
      "learning_rate": 9.600355522127174e-06,
      "loss": 1.3203207397460937,
      "step": 500
    },
    {
      "epoch": 0.517464424320828,
      "eval_loss": 1.2533953189849854,
      "eval_runtime": 178.6366,
      "eval_samples_per_second": 55.98,
      "eval_steps_per_second": 0.778,
      "step": 500
    },
    {
      "epoch": 0.5692108667529108,
      "grad_norm": 5.9375,
      "learning_rate": 9.481259637772251e-06,
      "loss": 1.3150039672851563,
      "step": 550
    },
    {
      "epoch": 0.6209573091849935,
      "grad_norm": 6.46875,
      "learning_rate": 9.347599809447331e-06,
      "loss": 1.3117788696289063,
      "step": 600
    },
    {
      "epoch": 0.6727037516170763,
      "grad_norm": 6.9375,
      "learning_rate": 9.199810427124904e-06,
      "loss": 1.3062660217285156,
      "step": 650
    },
    {
      "epoch": 0.7244501940491591,
      "grad_norm": 12.625,
      "learning_rate": 9.03837180135546e-06,
      "loss": 1.2984295654296876,
      "step": 700
    },
    {
      "epoch": 0.7761966364812419,
      "grad_norm": 6.21875,
      "learning_rate": 8.86380860227426e-06,
      "loss": 1.287723846435547,
      "step": 750
    },
    {
      "epoch": 0.7761966364812419,
      "eval_loss": 1.2287719249725342,
      "eval_runtime": 177.5762,
      "eval_samples_per_second": 56.314,
      "eval_steps_per_second": 0.783,
      "step": 750
    },
    {
      "epoch": 0.8279430789133247,
      "grad_norm": 8.3125,
      "learning_rate": 8.676688154440947e-06,
      "loss": 1.293363037109375,
      "step": 800
    },
    {
      "epoch": 0.8796895213454075,
      "grad_norm": 7.90625,
      "learning_rate": 8.477618593053693e-06,
      "loss": 1.2859170532226563,
      "step": 850
    },
    {
      "epoch": 0.9314359637774903,
      "grad_norm": 6.25,
      "learning_rate": 8.26724688753015e-06,
      "loss": 1.2893869018554687,
      "step": 900
    },
    {
      "epoch": 0.9831824062095731,
      "grad_norm": 5.21875,
      "learning_rate": 8.04625673887846e-06,
      "loss": 1.2764347076416016,
      "step": 950
    },
    {
      "epoch": 1.0341526520051747,
      "grad_norm": 4.90625,
      "learning_rate": 7.815366357691863e-06,
      "loss": 1.266595458984375,
      "step": 1000
    },
    {
      "epoch": 1.0341526520051747,
      "eval_loss": 1.2155812978744507,
      "eval_runtime": 177.1968,
      "eval_samples_per_second": 56.434,
      "eval_steps_per_second": 0.784,
      "step": 1000
    },
    {
      "epoch": 1.0858990944372575,
      "grad_norm": 6.9375,
      "learning_rate": 7.575326129988268e-06,
      "loss": 1.2680277252197265,
      "step": 1050
    },
    {
      "epoch": 1.1376455368693401,
      "grad_norm": 5.375,
      "learning_rate": 7.326916178480796e-06,
      "loss": 1.2629891204833985,
      "step": 1100
    },
    {
      "epoch": 1.189391979301423,
      "grad_norm": 8.75,
      "learning_rate": 7.070943827205048e-06,
      "loss": 1.2643635559082032,
      "step": 1150
    },
    {
      "epoch": 1.2411384217335057,
      "grad_norm": 11.9375,
      "learning_rate": 6.808240977742987e-06,
      "loss": 1.2576502227783204,
      "step": 1200
    },
    {
      "epoch": 1.2928848641655886,
      "grad_norm": 5.25,
      "learning_rate": 6.539661405570615e-06,
      "loss": 1.2591765594482422,
      "step": 1250
    },
    {
      "epoch": 1.2928848641655886,
      "eval_loss": 1.2061747312545776,
      "eval_runtime": 179.247,
      "eval_samples_per_second": 55.789,
      "eval_steps_per_second": 0.775,
      "step": 1250
    },
    {
      "epoch": 1.3446313065976714,
      "grad_norm": 5.03125,
      "learning_rate": 6.266077985316224e-06,
      "loss": 1.258944091796875,
      "step": 1300
    },
    {
      "epoch": 1.3963777490297542,
      "grad_norm": 7.90625,
      "learning_rate": 5.988379853947046e-06,
      "loss": 1.2634922790527343,
      "step": 1350
    },
    {
      "epoch": 1.448124191461837,
      "grad_norm": 7.59375,
      "learning_rate": 5.7074695211038385e-06,
      "loss": 1.2592449188232422,
      "step": 1400
    },
    {
      "epoch": 1.4998706338939198,
      "grad_norm": 5.90625,
      "learning_rate": 5.424259935974736e-06,
      "loss": 1.2600762939453125,
      "step": 1450
    },
    {
      "epoch": 1.5516170763260027,
      "grad_norm": 5.28125,
      "learning_rate": 5.139671520240915e-06,
      "loss": 1.2590911102294922,
      "step": 1500
    },
    {
      "epoch": 1.5516170763260027,
      "eval_loss": 1.201247215270996,
      "eval_runtime": 179.3983,
      "eval_samples_per_second": 55.742,
      "eval_steps_per_second": 0.775,
      "step": 1500
    },
    {
      "epoch": 1.6033635187580852,
      "grad_norm": 5.875,
      "learning_rate": 4.854629176736911e-06,
      "loss": 1.25985595703125,
      "step": 1550
    },
    {
      "epoch": 1.6551099611901683,
      "grad_norm": 6.09375,
      "learning_rate": 4.570059283547328e-06,
      "loss": 1.2547737121582032,
      "step": 1600
    },
    {
      "epoch": 1.7068564036222509,
      "grad_norm": 6.15625,
      "learning_rate": 4.286886683309053e-06,
      "loss": 1.2532640075683594,
      "step": 1650
    },
    {
      "epoch": 1.758602846054334,
      "grad_norm": 7.0625,
      "learning_rate": 4.0060316775036025e-06,
      "loss": 1.2515839385986327,
      "step": 1700
    },
    {
      "epoch": 1.8103492884864165,
      "grad_norm": 6.8125,
      "learning_rate": 3.728407035508089e-06,
      "loss": 1.2524348449707032,
      "step": 1750
    },
    {
      "epoch": 1.8103492884864165,
      "eval_loss": 1.198207974433899,
      "eval_runtime": 178.7391,
      "eval_samples_per_second": 55.947,
      "eval_steps_per_second": 0.778,
      "step": 1750
    },
    {
      "epoch": 1.8620957309184993,
      "grad_norm": 6.84375,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 1.2492080688476563,
      "step": 1800
    },
    {
      "epoch": 1.9138421733505822,
      "grad_norm": 6.59375,
      "learning_rate": 3.1864444952335823e-06,
      "loss": 1.2510016632080079,
      "step": 1850
    },
    {
      "epoch": 1.965588615782665,
      "grad_norm": 9.6875,
      "learning_rate": 2.9238679570873217e-06,
      "loss": 1.2530317687988282,
      "step": 1900
    },
    {
      "epoch": 2.0165588615782664,
      "grad_norm": 5.3125,
      "learning_rate": 2.66803877865493e-06,
      "loss": 1.254935760498047,
      "step": 1950
    },
    {
      "epoch": 2.0683053040103494,
      "grad_norm": 6.875,
      "learning_rate": 2.4197883962114164e-06,
      "loss": 1.2447527313232423,
      "step": 2000
    },
    {
      "epoch": 2.0683053040103494,
      "eval_loss": 1.1967895030975342,
      "eval_runtime": 179.8167,
      "eval_samples_per_second": 55.612,
      "eval_steps_per_second": 0.773,
      "step": 2000
    }
  ],
  "logging_steps": 50,
  "max_steps": 2901,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0500904956628828e+19,
  "train_batch_size": 18,
  "trial_name": null,
  "trial_params": null
}