{
  "best_global_step": 1100,
  "best_metric": 0.11870265007019043,
  "best_model_checkpoint": "./fine_tuning_results/checkpoint-1100",
  "epoch": 3.0,
  "eval_steps": 100,
  "global_step": 1131,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06637902422834384,
      "grad_norm": 5.728163719177246,
      "learning_rate": 0.0001411764705882353,
      "loss": 9.6583,
      "step": 25
    },
    {
      "epoch": 0.13275804845668768,
      "grad_norm": 11.366268157958984,
      "learning_rate": 0.00019990774875676051,
      "loss": 4.1605,
      "step": 50
    },
    {
      "epoch": 0.19913707268503153,
      "grad_norm": 4.53806209564209,
      "learning_rate": 0.0001993446074245224,
      "loss": 2.6629,
      "step": 75
    },
    {
      "epoch": 0.26551609691337535,
      "grad_norm": 3.489072799682617,
      "learning_rate": 0.0001982724572970665,
      "loss": 1.4258,
      "step": 100
    },
    {
      "epoch": 0.26551609691337535,
      "eval_loss": 0.2438984215259552,
      "eval_runtime": 410.0948,
      "eval_samples_per_second": 0.919,
      "eval_steps_per_second": 0.919,
      "step": 100
    },
    {
      "epoch": 0.3318951211417192,
      "grad_norm": 3.9948933124542236,
      "learning_rate": 0.00019669679171579118,
      "loss": 1.8838,
      "step": 125
    },
    {
      "epoch": 0.39827414537006306,
      "grad_norm": 2.3741233348846436,
      "learning_rate": 0.0001946256838681596,
      "loss": 1.2398,
      "step": 150
    },
    {
      "epoch": 0.4646531695984069,
      "grad_norm": 3.841597080230713,
      "learning_rate": 0.00019206974542336673,
      "loss": 1.6213,
      "step": 175
    },
    {
      "epoch": 0.5310321938267507,
      "grad_norm": 7.553189277648926,
      "learning_rate": 0.00018904207216166835,
      "loss": 1.0164,
      "step": 200
    },
    {
      "epoch": 0.5310321938267507,
      "eval_loss": 0.19300179183483124,
      "eval_runtime": 409.4453,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 200
    },
    {
      "epoch": 0.5974112180550946,
      "grad_norm": 4.948668956756592,
      "learning_rate": 0.00018555817687594984,
      "loss": 1.469,
      "step": 225
    },
    {
      "epoch": 0.6637902422834384,
      "grad_norm": 1.0395816564559937,
      "learning_rate": 0.00018163590988932402,
      "loss": 0.8695,
      "step": 250
    },
    {
      "epoch": 0.7301692665117823,
      "grad_norm": 3.0940940380096436,
      "learning_rate": 0.00017729536759600033,
      "loss": 1.5099,
      "step": 275
    },
    {
      "epoch": 0.7965482907401261,
      "grad_norm": 1.1767245531082153,
      "learning_rate": 0.000172558789494031,
      "loss": 0.8654,
      "step": 300
    },
    {
      "epoch": 0.7965482907401261,
      "eval_loss": 0.1677449345588684,
      "eval_runtime": 409.3224,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 300
    },
    {
      "epoch": 0.86292731496847,
      "grad_norm": 3.1094260215759277,
      "learning_rate": 0.00016745044423750448,
      "loss": 1.3233,
      "step": 325
    },
    {
      "epoch": 0.9293063391968138,
      "grad_norm": 3.251939296722412,
      "learning_rate": 0.00016199650529201683,
      "loss": 0.9238,
      "step": 350
    },
    {
      "epoch": 0.9956853634251577,
      "grad_norm": 5.835708141326904,
      "learning_rate": 0.00015622491683052125,
      "loss": 1.2047,
      "step": 375
    },
    {
      "epoch": 1.0610687022900764,
      "grad_norm": 2.8808865547180176,
      "learning_rate": 0.00015016525055666058,
      "loss": 1.1297,
      "step": 400
    },
    {
      "epoch": 1.0610687022900764,
      "eval_loss": 0.13486357033252716,
      "eval_runtime": 408.8177,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 400
    },
    {
      "epoch": 1.1274477265184202,
      "grad_norm": 4.205045223236084,
      "learning_rate": 0.00014384855418917312,
      "loss": 0.8077,
      "step": 425
    },
    {
      "epoch": 1.193826750746764,
      "grad_norm": 2.632800817489624,
      "learning_rate": 0.00013730719238368662,
      "loss": 1.1098,
      "step": 450
    },
    {
      "epoch": 1.2602057749751079,
      "grad_norm": 2.277942180633545,
      "learning_rate": 0.00013057468090696495,
      "loss": 0.8367,
      "step": 475
    },
    {
      "epoch": 1.3265847992034518,
      "grad_norm": 2.172236680984497,
      "learning_rate": 0.0001236855149132436,
      "loss": 1.0882,
      "step": 500
    },
    {
      "epoch": 1.3265847992034518,
      "eval_loss": 0.13017548620700836,
      "eval_runtime": 407.3514,
      "eval_samples_per_second": 0.925,
      "eval_steps_per_second": 0.925,
      "step": 500
    },
    {
      "epoch": 1.3929638234317956,
      "grad_norm": 3.1930105686187744,
      "learning_rate": 0.00011667499220250804,
      "loss": 0.8774,
      "step": 525
    },
    {
      "epoch": 1.4593428476601393,
      "grad_norm": 2.3342959880828857,
      "learning_rate": 0.00010957903236628266,
      "loss": 1.087,
      "step": 550
    },
    {
      "epoch": 1.5257218718884833,
      "grad_norm": 2.948979377746582,
      "learning_rate": 0.00010243399274756564,
      "loss": 0.8045,
      "step": 575
    },
    {
      "epoch": 1.592100896116827,
      "grad_norm": 2.6527910232543945,
      "learning_rate": 9.527648215787065e-05,
      "loss": 1.0704,
      "step": 600
    },
    {
      "epoch": 1.592100896116827,
      "eval_loss": 0.12589675188064575,
      "eval_runtime": 408.7252,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 600
    },
    {
      "epoch": 1.658479920345171,
      "grad_norm": 1.180399775505066,
      "learning_rate": 8.814317330582753e-05,
      "loss": 0.8398,
      "step": 625
    },
    {
      "epoch": 1.7248589445735147,
      "grad_norm": 2.150588274002075,
      "learning_rate": 8.107061489839499e-05,
      "loss": 1.0931,
      "step": 650
    },
    {
      "epoch": 1.7912379688018585,
      "grad_norm": 1.0235155820846558,
      "learning_rate": 7.409504437741722e-05,
      "loss": 0.8223,
      "step": 675
    },
    {
      "epoch": 1.8576169930302024,
      "grad_norm": 1.9297289848327637,
      "learning_rate": 6.725220225099911e-05,
      "loss": 0.9843,
      "step": 700
    },
    {
      "epoch": 1.8576169930302024,
      "eval_loss": 0.1218644380569458,
      "eval_runtime": 409.0035,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 700
    },
    {
      "epoch": 1.9239960172585464,
      "grad_norm": 3.172612190246582,
      "learning_rate": 6.0577148971005514e-05,
      "loss": 0.7813,
      "step": 725
    },
    {
      "epoch": 1.9903750414868902,
      "grad_norm": 2.5401861667633057,
      "learning_rate": 5.4104085294942505e-05,
      "loss": 0.7978,
      "step": 750
    },
    {
      "epoch": 2.0557583803518087,
      "grad_norm": 2.2131431102752686,
      "learning_rate": 4.786617705262746e-05,
      "loss": 0.8183,
      "step": 775
    },
    {
      "epoch": 2.122137404580153,
      "grad_norm": 4.252758026123047,
      "learning_rate": 4.189538521548524e-05,
      "loss": 0.706,
      "step": 800
    },
    {
      "epoch": 2.122137404580153,
      "eval_loss": 0.12229160219430923,
      "eval_runtime": 408.8022,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 800
    },
    {
      "epoch": 2.1885164288084966,
      "grad_norm": 2.4397575855255127,
      "learning_rate": 3.622230213913836e-05,
      "loss": 0.8344,
      "step": 825
    },
    {
      "epoch": 2.2548954530368404,
      "grad_norm": 1.8311405181884766,
      "learning_rate": 3.0875994818330955e-05,
      "loss": 0.6842,
      "step": 850
    },
    {
      "epoch": 2.321274477265184,
      "grad_norm": 2.2209343910217285,
      "learning_rate": 2.5883855957295056e-05,
      "loss": 0.8642,
      "step": 875
    },
    {
      "epoch": 2.387653501493528,
      "grad_norm": 1.0353050231933594,
      "learning_rate": 2.1271463618625986e-05,
      "loss": 0.6177,
      "step": 900
    },
    {
      "epoch": 2.387653501493528,
      "eval_loss": 0.12045033276081085,
      "eval_runtime": 409.5523,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 900
    },
    {
      "epoch": 2.454032525721872,
      "grad_norm": 2.0583794116973877,
      "learning_rate": 1.706245016977931e-05,
      "loss": 0.8361,
      "step": 925
    },
    {
      "epoch": 2.5204115499502158,
      "grad_norm": 2.405135154724121,
      "learning_rate": 1.3278381198663491e-05,
      "loss": 0.6861,
      "step": 950
    },
    {
      "epoch": 2.5867905741785595,
      "grad_norm": 2.8698174953460693,
      "learning_rate": 9.938645018725523e-06,
      "loss": 0.8369,
      "step": 975
    },
    {
      "epoch": 2.6531695984069037,
      "grad_norm": 2.2361364364624023,
      "learning_rate": 7.060353329667668e-06,
      "loss": 0.7129,
      "step": 1000
    },
    {
      "epoch": 2.6531695984069037,
      "eval_loss": 0.11899977177381516,
      "eval_runtime": 409.2689,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 1000
    },
    {
      "epoch": 2.719548622635247,
      "grad_norm": 2.688868522644043,
      "learning_rate": 4.658253542777991e-06,
      "loss": 0.8243,
      "step": 1025
    },
    {
      "epoch": 2.785927646863591,
      "grad_norm": 2.82051682472229,
      "learning_rate": 2.7446532200894106e-06,
      "loss": 0.6894,
      "step": 1050
    },
    {
      "epoch": 2.852306671091935,
      "grad_norm": 2.1398074626922607,
      "learning_rate": 1.3293570145169743e-06,
      "loss": 0.7943,
      "step": 1075
    },
    {
      "epoch": 2.9186856953202787,
      "grad_norm": 2.2756364345550537,
      "learning_rate": 4.196164340705577e-07,
      "loss": 0.6947,
      "step": 1100
    },
    {
      "epoch": 2.9186856953202787,
      "eval_loss": 0.11870265007019043,
      "eval_runtime": 409.6404,
      "eval_samples_per_second": 0.92,
      "eval_steps_per_second": 0.92,
      "step": 1100
    },
    {
      "epoch": 2.985064719548623,
      "grad_norm": 2.8848876953125,
      "learning_rate": 2.0092687534589703e-08,
      "loss": 0.7374,
      "step": 1125
    }
  ],
  "logging_steps": 25,
  "max_steps": 1131,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.5782738599655706e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}