{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 12032,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3324468085106383,
      "grad_norm": 0.3100210130214691,
      "learning_rate": 1.9168882978723405e-05,
      "loss": 0.0604,
      "step": 500
    },
    {
      "epoch": 0.3324468085106383,
      "eval_f1": 0.7222991689750693,
      "eval_loss": 0.051721904426813126,
      "eval_runtime": 17.61,
      "eval_samples_per_second": 85.349,
      "eval_steps_per_second": 10.676,
      "step": 500
    },
    {
      "epoch": 0.6648936170212766,
      "grad_norm": 1.4195618629455566,
      "learning_rate": 1.833776595744681e-05,
      "loss": 0.0531,
      "step": 1000
    },
    {
      "epoch": 0.6648936170212766,
      "eval_f1": 0.7194194885970975,
      "eval_loss": 0.05466139316558838,
      "eval_runtime": 17.5929,
      "eval_samples_per_second": 85.432,
      "eval_steps_per_second": 10.686,
      "step": 1000
    },
    {
      "epoch": 0.9973404255319149,
      "grad_norm": 1.0888991355895996,
      "learning_rate": 1.7506648936170213e-05,
      "loss": 0.0519,
      "step": 1500
    },
    {
      "epoch": 0.9973404255319149,
      "eval_f1": 0.737983375496928,
      "eval_loss": 0.04508286714553833,
      "eval_runtime": 17.6072,
      "eval_samples_per_second": 85.363,
      "eval_steps_per_second": 10.677,
      "step": 1500
    },
    {
      "epoch": 1.3297872340425532,
      "grad_norm": 0.06017958000302315,
      "learning_rate": 1.667553191489362e-05,
      "loss": 0.0319,
      "step": 2000
    },
    {
      "epoch": 1.3297872340425532,
      "eval_f1": 0.745473180731124,
      "eval_loss": 0.05267561972141266,
      "eval_runtime": 17.5974,
      "eval_samples_per_second": 85.41,
      "eval_steps_per_second": 10.683,
      "step": 2000
    },
    {
      "epoch": 1.6622340425531914,
      "grad_norm": 5.658984184265137,
      "learning_rate": 1.584441489361702e-05,
      "loss": 0.0361,
      "step": 2500
    },
    {
      "epoch": 1.6622340425531914,
      "eval_f1": 0.7482853223593965,
      "eval_loss": 0.05346440523862839,
      "eval_runtime": 17.5948,
      "eval_samples_per_second": 85.423,
      "eval_steps_per_second": 10.685,
      "step": 2500
    },
    {
      "epoch": 1.9946808510638299,
      "grad_norm": 2.0736632347106934,
      "learning_rate": 1.5013297872340426e-05,
      "loss": 0.0322,
      "step": 3000
    },
    {
      "epoch": 1.9946808510638299,
      "eval_f1": 0.7512027491408935,
      "eval_loss": 0.05298233404755592,
      "eval_runtime": 17.6025,
      "eval_samples_per_second": 85.385,
      "eval_steps_per_second": 10.68,
      "step": 3000
    },
    {
      "epoch": 2.327127659574468,
      "grad_norm": 0.005054426845163107,
      "learning_rate": 1.4182180851063831e-05,
      "loss": 0.0174,
      "step": 3500
    },
    {
      "epoch": 2.327127659574468,
      "eval_f1": 0.7525218560860794,
      "eval_loss": 0.07012591511011124,
      "eval_runtime": 17.5942,
      "eval_samples_per_second": 85.426,
      "eval_steps_per_second": 10.685,
      "step": 3500
    },
    {
      "epoch": 2.6595744680851063,
      "grad_norm": 0.475389301776886,
      "learning_rate": 1.3351063829787235e-05,
      "loss": 0.0218,
      "step": 4000
    },
    {
      "epoch": 2.6595744680851063,
      "eval_f1": 0.7464694014794889,
      "eval_loss": 0.0661526769399643,
      "eval_runtime": 17.6058,
      "eval_samples_per_second": 85.37,
      "eval_steps_per_second": 10.678,
      "step": 4000
    },
    {
      "epoch": 2.992021276595745,
      "grad_norm": 0.06960093975067139,
      "learning_rate": 1.2519946808510639e-05,
      "loss": 0.0215,
      "step": 4500
    },
    {
      "epoch": 2.992021276595745,
      "eval_f1": 0.7512864493996569,
      "eval_loss": 0.06172482669353485,
      "eval_runtime": 17.6025,
      "eval_samples_per_second": 85.385,
      "eval_steps_per_second": 10.68,
      "step": 4500
    },
    {
      "epoch": 3.324468085106383,
      "grad_norm": 0.00278457417152822,
      "learning_rate": 1.1688829787234044e-05,
      "loss": 0.0111,
      "step": 5000
    },
    {
      "epoch": 3.324468085106383,
      "eval_f1": 0.7572621035058431,
      "eval_loss": 0.08673229813575745,
      "eval_runtime": 17.6012,
      "eval_samples_per_second": 85.392,
      "eval_steps_per_second": 10.681,
      "step": 5000
    },
    {
      "epoch": 3.6569148936170213,
      "grad_norm": 0.18681606650352478,
      "learning_rate": 1.0857712765957446e-05,
      "loss": 0.0118,
      "step": 5500
    },
    {
      "epoch": 3.6569148936170213,
      "eval_f1": 0.7567567567567568,
      "eval_loss": 0.08375687897205353,
      "eval_runtime": 17.6087,
      "eval_samples_per_second": 85.356,
      "eval_steps_per_second": 10.677,
      "step": 5500
    },
    {
      "epoch": 3.9893617021276597,
      "grad_norm": 0.149847149848938,
      "learning_rate": 1.0026595744680852e-05,
      "loss": 0.0137,
      "step": 6000
    },
    {
      "epoch": 3.9893617021276597,
      "eval_f1": 0.7489075630252101,
      "eval_loss": 0.07563214004039764,
      "eval_runtime": 17.6009,
      "eval_samples_per_second": 85.393,
      "eval_steps_per_second": 10.681,
      "step": 6000
    },
    {
      "epoch": 4.321808510638298,
      "grad_norm": 0.003993071615695953,
      "learning_rate": 9.195478723404257e-06,
      "loss": 0.0067,
      "step": 6500
    },
    {
      "epoch": 4.321808510638298,
      "eval_f1": 0.747245409015025,
      "eval_loss": 0.09123753011226654,
      "eval_runtime": 17.6029,
      "eval_samples_per_second": 85.384,
      "eval_steps_per_second": 10.68,
      "step": 6500
    },
    {
      "epoch": 4.654255319148936,
      "grad_norm": 0.002212055493146181,
      "learning_rate": 8.36436170212766e-06,
      "loss": 0.0084,
      "step": 7000
    },
    {
      "epoch": 4.654255319148936,
      "eval_f1": 0.7503337783711616,
      "eval_loss": 0.08904670178890228,
      "eval_runtime": 17.6094,
      "eval_samples_per_second": 85.352,
      "eval_steps_per_second": 10.676,
      "step": 7000
    },
    {
      "epoch": 4.986702127659575,
      "grad_norm": 0.010206693783402443,
      "learning_rate": 7.5332446808510636e-06,
      "loss": 0.0066,
      "step": 7500
    },
    {
      "epoch": 4.986702127659575,
      "eval_f1": 0.7481629926519706,
      "eval_loss": 0.09713348001241684,
      "eval_runtime": 17.6091,
      "eval_samples_per_second": 85.354,
      "eval_steps_per_second": 10.676,
      "step": 7500
    },
    {
      "epoch": 5.319148936170213,
      "grad_norm": 0.0045317914336919785,
      "learning_rate": 6.702127659574469e-06,
      "loss": 0.005,
      "step": 8000
    },
    {
      "epoch": 5.319148936170213,
      "eval_f1": 0.7595865288429476,
      "eval_loss": 0.0952615961432457,
      "eval_runtime": 17.6003,
      "eval_samples_per_second": 85.396,
      "eval_steps_per_second": 10.682,
      "step": 8000
    },
    {
      "epoch": 5.651595744680851,
      "grad_norm": 0.0027383090928196907,
      "learning_rate": 5.871010638297873e-06,
      "loss": 0.0032,
      "step": 8500
    },
    {
      "epoch": 5.651595744680851,
      "eval_f1": 0.752435337588176,
      "eval_loss": 0.10410240292549133,
      "eval_runtime": 17.6009,
      "eval_samples_per_second": 85.393,
      "eval_steps_per_second": 10.681,
      "step": 8500
    },
    {
      "epoch": 5.98404255319149,
      "grad_norm": 13.284900665283203,
      "learning_rate": 5.039893617021277e-06,
      "loss": 0.0034,
      "step": 9000
    },
    {
      "epoch": 5.98404255319149,
      "eval_f1": 0.748834110592938,
      "eval_loss": 0.10824441909790039,
      "eval_runtime": 17.6038,
      "eval_samples_per_second": 85.379,
      "eval_steps_per_second": 10.68,
      "step": 9000
    },
    {
      "epoch": 6.316489361702128,
      "grad_norm": 0.04516634717583656,
      "learning_rate": 4.208776595744681e-06,
      "loss": 0.003,
      "step": 9500
    },
    {
      "epoch": 6.316489361702128,
      "eval_f1": 0.7523489932885906,
      "eval_loss": 0.1027175560593605,
      "eval_runtime": 17.6036,
      "eval_samples_per_second": 85.38,
      "eval_steps_per_second": 10.68,
      "step": 9500
    },
    {
      "epoch": 6.648936170212766,
      "grad_norm": 0.0026772848796099424,
      "learning_rate": 3.377659574468085e-06,
      "loss": 0.0024,
      "step": 10000
    },
    {
      "epoch": 6.648936170212766,
      "eval_f1": 0.7510829723425525,
      "eval_loss": 0.10764423757791519,
      "eval_runtime": 17.6186,
      "eval_samples_per_second": 85.308,
      "eval_steps_per_second": 10.671,
      "step": 10000
    },
    {
      "epoch": 6.9813829787234045,
      "grad_norm": 0.0036624702624976635,
      "learning_rate": 2.5465425531914894e-06,
      "loss": 0.0018,
      "step": 10500
    },
    {
      "epoch": 6.9813829787234045,
      "eval_f1": 0.7535845281760587,
      "eval_loss": 0.10941769182682037,
      "eval_runtime": 17.5982,
      "eval_samples_per_second": 85.406,
      "eval_steps_per_second": 10.683,
      "step": 10500
    },
    {
      "epoch": 7.3138297872340425,
      "grad_norm": 0.0005808394053019583,
      "learning_rate": 1.7154255319148937e-06,
      "loss": 0.0016,
      "step": 11000
    },
    {
      "epoch": 7.3138297872340425,
      "eval_f1": 0.7588510354041417,
      "eval_loss": 0.11245912313461304,
      "eval_runtime": 17.6057,
      "eval_samples_per_second": 85.37,
      "eval_steps_per_second": 10.678,
      "step": 11000
    },
    {
      "epoch": 7.6462765957446805,
      "grad_norm": 0.0013273729709908366,
      "learning_rate": 8.84308510638298e-07,
      "loss": 0.0009,
      "step": 11500
    },
    {
      "epoch": 7.6462765957446805,
      "eval_f1": 0.7561057209769153,
      "eval_loss": 0.11294491589069366,
      "eval_runtime": 17.6034,
      "eval_samples_per_second": 85.381,
      "eval_steps_per_second": 10.68,
      "step": 11500
    },
    {
      "epoch": 7.9787234042553195,
      "grad_norm": 0.002082614693790674,
      "learning_rate": 5.319148936170213e-08,
      "loss": 0.0007,
      "step": 12000
    },
    {
      "epoch": 7.9787234042553195,
      "eval_f1": 0.7553475935828877,
      "eval_loss": 0.11238062381744385,
      "eval_runtime": 17.6206,
      "eval_samples_per_second": 85.298,
      "eval_steps_per_second": 10.669,
      "step": 12000
    },
    {
      "epoch": 8.0,
      "step": 12032,
      "total_flos": 4.483356451828531e+16,
      "train_loss": 0.01690695836027481,
      "train_runtime": 2479.1734,
      "train_samples_per_second": 19.403,
      "train_steps_per_second": 4.853
    }
  ],
  "logging_steps": 500,
  "max_steps": 12032,
  "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": 4.483356451828531e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}