{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.333950903195924,
  "eval_steps": 8,
  "global_step": 90,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.014821676702176934,
      "eval_loss": 1.7156556844711304,
      "eval_runtime": 26.2876,
      "eval_samples_per_second": 8.673,
      "eval_steps_per_second": 4.337,
      "step": 1
    },
    {
      "epoch": 0.0444650301065308,
      "grad_norm": 2.3580570220947266,
      "learning_rate": 3e-05,
      "loss": 1.7416,
      "step": 3
    },
    {
      "epoch": 0.0889300602130616,
      "grad_norm": 1.4993271827697754,
      "learning_rate": 6e-05,
      "loss": 1.5326,
      "step": 6
    },
    {
      "epoch": 0.11857341361741547,
      "eval_loss": 0.9888727068901062,
      "eval_runtime": 26.3261,
      "eval_samples_per_second": 8.661,
      "eval_steps_per_second": 4.33,
      "step": 8
    },
    {
      "epoch": 0.1333950903195924,
      "grad_norm": 1.521828532218933,
      "learning_rate": 9e-05,
      "loss": 1.0888,
      "step": 9
    },
    {
      "epoch": 0.1778601204261232,
      "grad_norm": 1.2248362302780151,
      "learning_rate": 0.00012,
      "loss": 0.8213,
      "step": 12
    },
    {
      "epoch": 0.222325150532654,
      "grad_norm": 1.055378794670105,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.7635,
      "step": 15
    },
    {
      "epoch": 0.23714682723483094,
      "eval_loss": 0.5862013101577759,
      "eval_runtime": 26.3342,
      "eval_samples_per_second": 8.658,
      "eval_steps_per_second": 4.329,
      "step": 16
    },
    {
      "epoch": 0.2667901806391848,
      "grad_norm": 1.1668015718460083,
      "learning_rate": 0.00018,
      "loss": 0.6722,
      "step": 18
    },
    {
      "epoch": 0.3112552107457156,
      "grad_norm": 1.0489511489868164,
      "learning_rate": 0.00019989930665413147,
      "loss": 0.6261,
      "step": 21
    },
    {
      "epoch": 0.3557202408522464,
      "grad_norm": 0.8146524429321289,
      "learning_rate": 0.00019839295885986296,
      "loss": 0.6004,
      "step": 24
    },
    {
      "epoch": 0.3557202408522464,
      "eval_loss": 0.5033943057060242,
      "eval_runtime": 26.4413,
      "eval_samples_per_second": 8.623,
      "eval_steps_per_second": 4.311,
      "step": 24
    },
    {
      "epoch": 0.4001852709587772,
      "grad_norm": 0.8943031430244446,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.5921,
      "step": 27
    },
    {
      "epoch": 0.444650301065308,
      "grad_norm": 0.7065706849098206,
      "learning_rate": 0.0001900968867902419,
      "loss": 0.5142,
      "step": 30
    },
    {
      "epoch": 0.4742936544696619,
      "eval_loss": 0.4665619134902954,
      "eval_runtime": 26.3696,
      "eval_samples_per_second": 8.646,
      "eval_steps_per_second": 4.323,
      "step": 32
    },
    {
      "epoch": 0.4891153311718388,
      "grad_norm": 0.82402104139328,
      "learning_rate": 0.00018345732537213027,
      "loss": 0.5384,
      "step": 33
    },
    {
      "epoch": 0.5335803612783696,
      "grad_norm": 0.7770050764083862,
      "learning_rate": 0.00017530714660036112,
      "loss": 0.5215,
      "step": 36
    },
    {
      "epoch": 0.5780453913849004,
      "grad_norm": 0.6626347899436951,
      "learning_rate": 0.00016579387259397127,
      "loss": 0.5285,
      "step": 39
    },
    {
      "epoch": 0.5928670680870773,
      "eval_loss": 0.45161083340644836,
      "eval_runtime": 26.3307,
      "eval_samples_per_second": 8.659,
      "eval_steps_per_second": 4.33,
      "step": 40
    },
    {
      "epoch": 0.6225104214914312,
      "grad_norm": 0.6447142362594604,
      "learning_rate": 0.00015508969814521025,
      "loss": 0.5811,
      "step": 42
    },
    {
      "epoch": 0.666975451597962,
      "grad_norm": 0.6098252534866333,
      "learning_rate": 0.00014338837391175582,
      "loss": 0.4708,
      "step": 45
    },
    {
      "epoch": 0.7114404817044928,
      "grad_norm": 0.669792890548706,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.4997,
      "step": 48
    },
    {
      "epoch": 0.7114404817044928,
      "eval_loss": 0.43807509541511536,
      "eval_runtime": 26.3421,
      "eval_samples_per_second": 8.655,
      "eval_steps_per_second": 4.328,
      "step": 48
    },
    {
      "epoch": 0.7559055118110236,
      "grad_norm": 0.6415955424308777,
      "learning_rate": 0.00011785568947986367,
      "loss": 0.5092,
      "step": 51
    },
    {
      "epoch": 0.8003705419175544,
      "grad_norm": 0.61659836769104,
      "learning_rate": 0.00010448648303505151,
      "loss": 0.5257,
      "step": 54
    },
    {
      "epoch": 0.8300138953219083,
      "eval_loss": 0.4247357249259949,
      "eval_runtime": 26.342,
      "eval_samples_per_second": 8.655,
      "eval_steps_per_second": 4.328,
      "step": 56
    },
    {
      "epoch": 0.8448355720240852,
      "grad_norm": 0.6660127639770508,
      "learning_rate": 9.103606910965666e-05,
      "loss": 0.5072,
      "step": 57
    },
    {
      "epoch": 0.889300602130616,
      "grad_norm": 0.6248095035552979,
      "learning_rate": 7.774790660436858e-05,
      "loss": 0.485,
      "step": 60
    },
    {
      "epoch": 0.9337656322371468,
      "grad_norm": 0.6779568195343018,
      "learning_rate": 6.486251759186572e-05,
      "loss": 0.4735,
      "step": 63
    },
    {
      "epoch": 0.9485873089393237,
      "eval_loss": 0.41895875334739685,
      "eval_runtime": 26.3269,
      "eval_samples_per_second": 8.66,
      "eval_steps_per_second": 4.33,
      "step": 64
    },
    {
      "epoch": 0.9782306623436776,
      "grad_norm": 0.5692128539085388,
      "learning_rate": 5.261313375270014e-05,
      "loss": 0.482,
      "step": 66
    },
    {
      "epoch": 1.0226956924502084,
      "grad_norm": 0.4736873507499695,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.4658,
      "step": 69
    },
    {
      "epoch": 1.0671607225567392,
      "grad_norm": 0.6735922694206238,
      "learning_rate": 3.089373510131354e-05,
      "loss": 0.4016,
      "step": 72
    },
    {
      "epoch": 1.0671607225567392,
      "eval_loss": 0.41547104716300964,
      "eval_runtime": 26.3371,
      "eval_samples_per_second": 8.657,
      "eval_steps_per_second": 4.329,
      "step": 72
    },
    {
      "epoch": 1.11162575266327,
      "grad_norm": 0.5336905121803284,
      "learning_rate": 2.181685175319702e-05,
      "loss": 0.3519,
      "step": 75
    },
    {
      "epoch": 1.1560907827698008,
      "grad_norm": 0.5846615433692932,
      "learning_rate": 1.415512063981339e-05,
      "loss": 0.3481,
      "step": 78
    },
    {
      "epoch": 1.1857341361741547,
      "eval_loss": 0.4153323173522949,
      "eval_runtime": 26.3347,
      "eval_samples_per_second": 8.658,
      "eval_steps_per_second": 4.329,
      "step": 80
    },
    {
      "epoch": 1.2005558128763316,
      "grad_norm": 0.49688470363616943,
      "learning_rate": 8.047222744854943e-06,
      "loss": 0.3458,
      "step": 81
    },
    {
      "epoch": 1.2450208429828624,
      "grad_norm": 0.6959128379821777,
      "learning_rate": 3.6037139304146762e-06,
      "loss": 0.354,
      "step": 84
    },
    {
      "epoch": 1.2894858730893932,
      "grad_norm": 0.6246429085731506,
      "learning_rate": 9.0502382320653e-07,
      "loss": 0.3348,
      "step": 87
    },
    {
      "epoch": 1.3043075497915702,
      "eval_loss": 0.4123484492301941,
      "eval_runtime": 26.3287,
      "eval_samples_per_second": 8.66,
      "eval_steps_per_second": 4.33,
      "step": 88
    },
    {
      "epoch": 1.333950903195924,
      "grad_norm": 0.6873133778572083,
      "learning_rate": 0.0,
      "loss": 0.3512,
      "step": 90
    }
  ],
  "logging_steps": 3,
  "max_steps": 90,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 8,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.587618286712914e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}