{
  "best_global_step": 3230,
  "best_metric": 1.2906097173690796,
  "best_model_checkpoint": "/content/drive/MyDrive/mn_cpt_v3_ckpts_round2/checkpoint-3230",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 3230,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.030959752321981424,
      "grad_norm": 1.7435482740402222,
      "learning_rate": 9.977069965464679e-06,
      "loss": 1.3856842041015625,
      "step": 100
    },
    {
      "epoch": 0.06191950464396285,
      "grad_norm": 4.383045673370361,
      "learning_rate": 9.90756854118645e-06,
      "loss": 1.3523806762695312,
      "step": 200
    },
    {
      "epoch": 0.09287925696594428,
      "grad_norm": 10.109758377075195,
      "learning_rate": 9.79215035650482e-06,
      "loss": 1.3374905395507812,
      "step": 300
    },
    {
      "epoch": 0.1238390092879257,
      "grad_norm": 2.4500350952148438,
      "learning_rate": 9.631906415918705e-06,
      "loss": 1.3301382446289063,
      "step": 400
    },
    {
      "epoch": 0.15479876160990713,
      "grad_norm": 2.826953887939453,
      "learning_rate": 9.428351444864008e-06,
      "loss": 1.3228797912597656,
      "step": 500
    },
    {
      "epoch": 0.15479876160990713,
      "eval_loss": 1.3230881690979004,
      "eval_runtime": 36.129,
      "eval_samples_per_second": 11.071,
      "eval_steps_per_second": 1.384,
      "step": 500
    },
    {
      "epoch": 0.18575851393188855,
      "grad_norm": 2.2888803482055664,
      "learning_rate": 9.183409571586292e-06,
      "loss": 1.317882080078125,
      "step": 600
    },
    {
      "epoch": 0.21671826625386997,
      "grad_norm": 3.3586924076080322,
      "learning_rate": 8.899396139083332e-06,
      "loss": 1.3019407653808595,
      "step": 700
    },
    {
      "epoch": 0.2476780185758514,
      "grad_norm": 4.1671576499938965,
      "learning_rate": 8.578995819042417e-06,
      "loss": 1.307056884765625,
      "step": 800
    },
    {
      "epoch": 0.2786377708978328,
      "grad_norm": 3.3320398330688477,
      "learning_rate": 8.225237234653058e-06,
      "loss": 1.312891082763672,
      "step": 900
    },
    {
      "epoch": 0.30959752321981426,
      "grad_norm": 3.507591485977173,
      "learning_rate": 7.841464332176092e-06,
      "loss": 1.3050839233398437,
      "step": 1000
    },
    {
      "epoch": 0.30959752321981426,
      "eval_loss": 1.3081670999526978,
      "eval_runtime": 26.5452,
      "eval_samples_per_second": 15.069,
      "eval_steps_per_second": 1.884,
      "step": 1000
    },
    {
      "epoch": 0.34055727554179566,
      "grad_norm": 4.539306640625,
      "learning_rate": 7.431304771882899e-06,
      "loss": 1.2991960144042969,
      "step": 1100
    },
    {
      "epoch": 0.3715170278637771,
      "grad_norm": 3.7326371669769287,
      "learning_rate": 6.998635637153249e-06,
      "loss": 1.2984297180175781,
      "step": 1200
    },
    {
      "epoch": 0.4024767801857585,
      "grad_norm": 3.579521894454956,
      "learning_rate": 6.547546785870751e-06,
      "loss": 1.3055659484863282,
      "step": 1300
    },
    {
      "epoch": 0.43343653250773995,
      "grad_norm": 3.2823829650878906,
      "learning_rate": 6.082302190541269e-06,
      "loss": 1.294408721923828,
      "step": 1400
    },
    {
      "epoch": 0.46439628482972134,
      "grad_norm": 9.245798110961914,
      "learning_rate": 5.607299632571611e-06,
      "loss": 1.295184326171875,
      "step": 1500
    },
    {
      "epoch": 0.46439628482972134,
      "eval_loss": 1.3002694845199585,
      "eval_runtime": 26.6265,
      "eval_samples_per_second": 15.023,
      "eval_steps_per_second": 1.878,
      "step": 1500
    },
    {
      "epoch": 0.4953560371517028,
      "grad_norm": 3.497911214828491,
      "learning_rate": 5.127029131703213e-06,
      "loss": 1.2944673156738282,
      "step": 1600
    },
    {
      "epoch": 0.5263157894736842,
      "grad_norm": 4.088214874267578,
      "learning_rate": 4.646030503551686e-06,
      "loss": 1.2922651672363281,
      "step": 1700
    },
    {
      "epoch": 0.5572755417956656,
      "grad_norm": 3.931842565536499,
      "learning_rate": 4.168850446444794e-06,
      "loss": 1.29747802734375,
      "step": 1800
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 10.10274887084961,
      "learning_rate": 3.6999995632007324e-06,
      "loss": 1.2896998596191407,
      "step": 1900
    },
    {
      "epoch": 0.6191950464396285,
      "grad_norm": 5.449999809265137,
      "learning_rate": 3.243909724103639e-06,
      "loss": 1.2893682861328124,
      "step": 2000
    },
    {
      "epoch": 0.6191950464396285,
      "eval_loss": 1.2946323156356812,
      "eval_runtime": 26.668,
      "eval_samples_per_second": 14.999,
      "eval_steps_per_second": 1.875,
      "step": 2000
    },
    {
      "epoch": 0.6501547987616099,
      "grad_norm": 5.21806526184082,
      "learning_rate": 2.8048921741079935e-06,
      "loss": 1.2730223083496093,
      "step": 2100
    },
    {
      "epoch": 0.6811145510835913,
      "grad_norm": 5.030884742736816,
      "learning_rate": 2.38709678026871e-06,
      "loss": 1.3017454528808594,
      "step": 2200
    },
    {
      "epoch": 0.7120743034055728,
      "grad_norm": 4.097794055938721,
      "learning_rate": 1.994472804615526e-06,
      "loss": 1.2895455932617188,
      "step": 2300
    },
    {
      "epoch": 0.7430340557275542,
      "grad_norm": 4.602784633636475,
      "learning_rate": 1.6307315732709309e-06,
      "loss": 1.283674774169922,
      "step": 2400
    },
    {
      "epoch": 0.7739938080495357,
      "grad_norm": 7.153234004974365,
      "learning_rate": 1.2993113946863443e-06,
      "loss": 1.2933894348144532,
      "step": 2500
    },
    {
      "epoch": 0.7739938080495357,
      "eval_loss": 1.2917898893356323,
      "eval_runtime": 26.6306,
      "eval_samples_per_second": 15.02,
      "eval_steps_per_second": 1.878,
      "step": 2500
    },
    {
      "epoch": 0.804953560371517,
      "grad_norm": 6.6482014656066895,
      "learning_rate": 1.0033450586111783e-06,
      "loss": 1.289451904296875,
      "step": 2600
    },
    {
      "epoch": 0.8359133126934984,
      "grad_norm": 13.014872550964355,
      "learning_rate": 7.456302230148107e-07,
      "loss": 1.2901728820800782,
      "step": 2700
    },
    {
      "epoch": 0.8668730650154799,
      "grad_norm": 5.961217880249023,
      "learning_rate": 5.286029688826459e-07,
      "loss": 1.2795370483398438,
      "step": 2800
    },
    {
      "epoch": 0.8978328173374613,
      "grad_norm": 4.641078472137451,
      "learning_rate": 3.543147728628245e-07,
      "loss": 1.2775550842285157,
      "step": 2900
    },
    {
      "epoch": 0.9287925696594427,
      "grad_norm": 4.651384353637695,
      "learning_rate": 2.2441311543244762e-07,
      "loss": 1.2817430114746093,
      "step": 3000
    },
    {
      "epoch": 0.9287925696594427,
      "eval_loss": 1.2906177043914795,
      "eval_runtime": 26.6781,
      "eval_samples_per_second": 14.994,
      "eval_steps_per_second": 1.874,
      "step": 3000
    },
    {
      "epoch": 0.9597523219814241,
      "grad_norm": 4.312687873840332,
      "learning_rate": 1.4012590788698408e-07,
      "loss": 1.2772533416748046,
      "step": 3100
    },
    {
      "epoch": 0.9907120743034056,
      "grad_norm": 4.400423526763916,
      "learning_rate": 1.022498853586551e-07,
      "loss": 1.2894178771972655,
      "step": 3200
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.2906097173690796,
      "eval_runtime": 26.6681,
      "eval_samples_per_second": 14.999,
      "eval_steps_per_second": 1.875,
      "step": 3230
    }
  ],
  "logging_steps": 100,
  "max_steps": 3230,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 2.5157373092010117e+18,
  "train_batch_size": 45,
  "trial_name": null,
  "trial_params": null
}