{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 200,
  "global_step": 878,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04555808656036447,
      "grad_norm": 14.260233879089355,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.6404,
      "step": 20
    },
    {
      "epoch": 0.09111617312072894,
      "grad_norm": 13.856605529785156,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.6072,
      "step": 40
    },
    {
      "epoch": 0.1366742596810934,
      "grad_norm": 10.69395637512207,
      "learning_rate": 1.5e-06,
      "loss": 0.4863,
      "step": 60
    },
    {
      "epoch": 0.18223234624145787,
      "grad_norm": 6.665396213531494,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.3266,
      "step": 80
    },
    {
      "epoch": 0.22779043280182232,
      "grad_norm": 2.674818754196167,
      "learning_rate": 2.5e-06,
      "loss": 0.1972,
      "step": 100
    },
    {
      "epoch": 0.2733485193621868,
      "grad_norm": 0.46903184056282043,
      "learning_rate": 3e-06,
      "loss": 0.0999,
      "step": 120
    },
    {
      "epoch": 0.31890660592255127,
      "grad_norm": 0.6415174603462219,
      "learning_rate": 3.5e-06,
      "loss": 0.0698,
      "step": 140
    },
    {
      "epoch": 0.36446469248291574,
      "grad_norm": 0.3701508343219757,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0656,
      "step": 160
    },
    {
      "epoch": 0.41002277904328016,
      "grad_norm": 0.4025450348854065,
      "learning_rate": 4.5e-06,
      "loss": 0.0574,
      "step": 180
    },
    {
      "epoch": 0.45558086560364464,
      "grad_norm": 0.5695033073425293,
      "learning_rate": 5e-06,
      "loss": 0.0562,
      "step": 200
    },
    {
      "epoch": 0.45558086560364464,
      "eval_accuracy": 0.8391241361293846,
      "eval_f1": 0.8391241361293846,
      "eval_f1_marco": 0.8249262659790968,
      "eval_loss": 0.056919749826192856,
      "eval_negative_f1": 0.8747828015823136,
      "eval_positive_f1": 0.7750697303758799,
      "eval_precision": 0.8391241361293846,
      "eval_recall": 0.8391241361293846,
      "eval_runtime": 9.3166,
      "eval_samples_per_second": 79.106,
      "eval_steps_per_second": 1.288,
      "step": 200
    },
    {
      "epoch": 0.5011389521640092,
      "grad_norm": 0.527057945728302,
      "learning_rate": 4.852507374631269e-06,
      "loss": 0.0622,
      "step": 220
    },
    {
      "epoch": 0.5466970387243736,
      "grad_norm": 0.47804608941078186,
      "learning_rate": 4.705014749262537e-06,
      "loss": 0.0527,
      "step": 240
    },
    {
      "epoch": 0.592255125284738,
      "grad_norm": 0.3202660381793976,
      "learning_rate": 4.557522123893805e-06,
      "loss": 0.0555,
      "step": 260
    },
    {
      "epoch": 0.6378132118451025,
      "grad_norm": 0.40933147072792053,
      "learning_rate": 4.410029498525074e-06,
      "loss": 0.0563,
      "step": 280
    },
    {
      "epoch": 0.683371298405467,
      "grad_norm": 0.4464198648929596,
      "learning_rate": 4.2625368731563425e-06,
      "loss": 0.054,
      "step": 300
    },
    {
      "epoch": 0.7289293849658315,
      "grad_norm": 0.5946183204650879,
      "learning_rate": 4.115044247787611e-06,
      "loss": 0.0543,
      "step": 320
    },
    {
      "epoch": 0.7744874715261959,
      "grad_norm": 0.8823751211166382,
      "learning_rate": 3.967551622418879e-06,
      "loss": 0.0552,
      "step": 340
    },
    {
      "epoch": 0.8200455580865603,
      "grad_norm": 0.29086050391197205,
      "learning_rate": 3.820058997050148e-06,
      "loss": 0.0556,
      "step": 360
    },
    {
      "epoch": 0.8656036446469249,
      "grad_norm": 0.36109957098960876,
      "learning_rate": 3.6725663716814163e-06,
      "loss": 0.0547,
      "step": 380
    },
    {
      "epoch": 0.9111617312072893,
      "grad_norm": 0.357105553150177,
      "learning_rate": 3.5250737463126845e-06,
      "loss": 0.054,
      "step": 400
    },
    {
      "epoch": 0.9111617312072893,
      "eval_accuracy": 0.857600873963949,
      "eval_f1": 0.857600873963949,
      "eval_f1_marco": 0.8454311469742184,
      "eval_loss": 0.05031890422105789,
      "eval_negative_f1": 0.8888023441267016,
      "eval_positive_f1": 0.8020599498217351,
      "eval_precision": 0.857600873963949,
      "eval_recall": 0.857600873963949,
      "eval_runtime": 9.081,
      "eval_samples_per_second": 81.158,
      "eval_steps_per_second": 1.321,
      "step": 400
    },
    {
      "epoch": 0.9567198177676538,
      "grad_norm": 0.3978097438812256,
      "learning_rate": 3.3775811209439528e-06,
      "loss": 0.0513,
      "step": 420
    },
    {
      "epoch": 1.0022779043280183,
      "grad_norm": 0.327373206615448,
      "learning_rate": 3.2300884955752214e-06,
      "loss": 0.0527,
      "step": 440
    },
    {
      "epoch": 1.0478359908883828,
      "grad_norm": 0.39979490637779236,
      "learning_rate": 3.08259587020649e-06,
      "loss": 0.0474,
      "step": 460
    },
    {
      "epoch": 1.0933940774487472,
      "grad_norm": 0.37922340631484985,
      "learning_rate": 2.935103244837758e-06,
      "loss": 0.0501,
      "step": 480
    },
    {
      "epoch": 1.1389521640091116,
      "grad_norm": 0.4099065363407135,
      "learning_rate": 2.7876106194690266e-06,
      "loss": 0.0461,
      "step": 500
    },
    {
      "epoch": 1.184510250569476,
      "grad_norm": 0.3328123390674591,
      "learning_rate": 2.6401179941002952e-06,
      "loss": 0.048,
      "step": 520
    },
    {
      "epoch": 1.2300683371298406,
      "grad_norm": 0.647693932056427,
      "learning_rate": 2.4926253687315635e-06,
      "loss": 0.0495,
      "step": 540
    },
    {
      "epoch": 1.275626423690205,
      "grad_norm": 0.6742229461669922,
      "learning_rate": 2.345132743362832e-06,
      "loss": 0.0496,
      "step": 560
    },
    {
      "epoch": 1.3211845102505695,
      "grad_norm": 0.2932397425174713,
      "learning_rate": 2.1976401179941004e-06,
      "loss": 0.0479,
      "step": 580
    },
    {
      "epoch": 1.366742596810934,
      "grad_norm": 0.32655268907546997,
      "learning_rate": 2.050147492625369e-06,
      "loss": 0.0472,
      "step": 600
    },
    {
      "epoch": 1.366742596810934,
      "eval_accuracy": 0.8680504429192296,
      "eval_f1": 0.8680504429192296,
      "eval_f1_marco": 0.853667569896885,
      "eval_loss": 0.04778573289513588,
      "eval_negative_f1": 0.8995443697114341,
      "eval_positive_f1": 0.8077907700823358,
      "eval_precision": 0.8680504429192296,
      "eval_recall": 0.8680504429192296,
      "eval_runtime": 9.6373,
      "eval_samples_per_second": 76.473,
      "eval_steps_per_second": 1.245,
      "step": 600
    },
    {
      "epoch": 1.4123006833712983,
      "grad_norm": 0.3308158814907074,
      "learning_rate": 1.9026548672566373e-06,
      "loss": 0.0455,
      "step": 620
    },
    {
      "epoch": 1.4578587699316627,
      "grad_norm": 0.4326237738132477,
      "learning_rate": 1.7551622418879058e-06,
      "loss": 0.0476,
      "step": 640
    },
    {
      "epoch": 1.5034168564920274,
      "grad_norm": 0.9873289465904236,
      "learning_rate": 1.607669616519174e-06,
      "loss": 0.0469,
      "step": 660
    },
    {
      "epoch": 1.5489749430523918,
      "grad_norm": 0.4288870096206665,
      "learning_rate": 1.4601769911504427e-06,
      "loss": 0.0459,
      "step": 680
    },
    {
      "epoch": 1.5945330296127562,
      "grad_norm": 0.4720146358013153,
      "learning_rate": 1.312684365781711e-06,
      "loss": 0.0494,
      "step": 700
    },
    {
      "epoch": 1.6400911161731209,
      "grad_norm": 0.6934795379638672,
      "learning_rate": 1.1651917404129796e-06,
      "loss": 0.05,
      "step": 720
    },
    {
      "epoch": 1.6856492027334853,
      "grad_norm": 0.3552420735359192,
      "learning_rate": 1.017699115044248e-06,
      "loss": 0.0499,
      "step": 740
    },
    {
      "epoch": 1.7312072892938497,
      "grad_norm": 0.32909882068634033,
      "learning_rate": 8.702064896755164e-07,
      "loss": 0.0485,
      "step": 760
    },
    {
      "epoch": 1.7767653758542141,
      "grad_norm": 0.2789745628833771,
      "learning_rate": 7.227138643067848e-07,
      "loss": 0.0468,
      "step": 780
    },
    {
      "epoch": 1.8223234624145785,
      "grad_norm": 0.3232771158218384,
      "learning_rate": 5.752212389380532e-07,
      "loss": 0.0481,
      "step": 800
    },
    {
      "epoch": 1.8223234624145785,
      "eval_accuracy": 0.8677417056546417,
      "eval_f1": 0.8677417056546417,
      "eval_f1_marco": 0.8536354633333805,
      "eval_loss": 0.04745788872241974,
      "eval_negative_f1": 0.8990739230504359,
      "eval_positive_f1": 0.8081970036163251,
      "eval_precision": 0.8677417056546417,
      "eval_recall": 0.8677417056546417,
      "eval_runtime": 8.8278,
      "eval_samples_per_second": 83.487,
      "eval_steps_per_second": 1.359,
      "step": 800
    },
    {
      "epoch": 1.867881548974943,
      "grad_norm": 0.4403178095817566,
      "learning_rate": 4.277286135693216e-07,
      "loss": 0.0502,
      "step": 820
    },
    {
      "epoch": 1.9134396355353074,
      "grad_norm": 0.4974011182785034,
      "learning_rate": 2.8023598820059e-07,
      "loss": 0.0453,
      "step": 840
    },
    {
      "epoch": 1.958997722095672,
      "grad_norm": 0.3315702974796295,
      "learning_rate": 1.327433628318584e-07,
      "loss": 0.0494,
      "step": 860
    }
  ],
  "logging_steps": 20,
  "max_steps": 878,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 1.952805421392077e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}