{
  "best_metric": 0.8502547144889832,
  "best_model_checkpoint": "data/smollm2-1.7B-sft\\checkpoint-142",
  "epoch": 1.9984,
  "eval_steps": 500,
  "global_step": 312,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.035398230088495575,
      "grad_norm": 8.5625,
      "learning_rate": 0.0001724137931034483,
      "loss": 2.2924,
      "mean_token_accuracy": 0.5223099514842033,
      "step": 5
    },
    {
      "epoch": 0.07079646017699115,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003448275862068966,
      "loss": 1.07,
      "mean_token_accuracy": 0.7215210050344467,
      "step": 10
    },
    {
      "epoch": 0.10619469026548672,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0005172413793103448,
      "loss": 0.8398,
      "mean_token_accuracy": 0.7689230531454087,
      "step": 15
    },
    {
      "epoch": 0.1415929203539823,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0006896551724137932,
      "loss": 0.8466,
      "mean_token_accuracy": 0.7690734684467315,
      "step": 20
    },
    {
      "epoch": 0.17699115044247787,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0008620689655172414,
      "loss": 0.8315,
      "mean_token_accuracy": 0.7698834031820297,
      "step": 25
    },
    {
      "epoch": 0.21238938053097345,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000999961452773888,
      "loss": 0.8657,
      "mean_token_accuracy": 0.7694115966558457,
      "step": 30
    },
    {
      "epoch": 0.24778761061946902,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0009986129238305635,
      "loss": 0.9508,
      "mean_token_accuracy": 0.7505242705345154,
      "step": 35
    },
    {
      "epoch": 0.2831858407079646,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0009953429730181654,
      "loss": 0.9422,
      "mean_token_accuracy": 0.7508314579725266,
      "step": 40
    },
    {
      "epoch": 0.3185840707964602,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0009901642012034213,
      "loss": 0.9568,
      "mean_token_accuracy": 0.7477447241544724,
      "step": 45
    },
    {
      "epoch": 0.35398230088495575,
      "grad_norm": 0.75390625,
      "learning_rate": 0.0009830965649597454,
      "loss": 0.9577,
      "mean_token_accuracy": 0.7479827046394348,
      "step": 50
    },
    {
      "epoch": 0.3893805309734513,
      "grad_norm": 0.69921875,
      "learning_rate": 0.0009741672996639047,
      "loss": 0.9318,
      "mean_token_accuracy": 0.7587000578641891,
      "step": 55
    },
    {
      "epoch": 0.4247787610619469,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0009634108145435665,
      "loss": 0.9259,
      "mean_token_accuracy": 0.7593102544546128,
      "step": 60
    },
    {
      "epoch": 0.46017699115044247,
      "grad_norm": 0.7109375,
      "learning_rate": 0.0009508685600801703,
      "loss": 0.8877,
      "mean_token_accuracy": 0.768345057964325,
      "step": 65
    },
    {
      "epoch": 0.49557522123893805,
      "grad_norm": 0.66015625,
      "learning_rate": 0.0009365888682780861,
      "loss": 0.9253,
      "mean_token_accuracy": 0.7625227481126785,
      "step": 70
    },
    {
      "epoch": 0.5309734513274337,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0009206267664155906,
      "loss": 0.912,
      "mean_token_accuracy": 0.7615309178829193,
      "step": 75
    },
    {
      "epoch": 0.5663716814159292,
      "grad_norm": 0.67578125,
      "learning_rate": 0.0009030437649953789,
      "loss": 0.9242,
      "mean_token_accuracy": 0.7563268959522247,
      "step": 80
    },
    {
      "epoch": 0.6017699115044248,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0008839076207117484,
      "loss": 0.8975,
      "mean_token_accuracy": 0.7675409287214279,
      "step": 85
    },
    {
      "epoch": 0.6371681415929203,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0008632920753478719,
      "loss": 0.8901,
      "mean_token_accuracy": 0.7678898781538009,
      "step": 90
    },
    {
      "epoch": 0.672566371681416,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0008412765716093271,
      "loss": 0.8868,
      "mean_token_accuracy": 0.7646951735019684,
      "step": 95
    },
    {
      "epoch": 0.7079646017699115,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0008179459469889268,
      "loss": 0.9314,
      "mean_token_accuracy": 0.7571736365556717,
      "step": 100
    },
    {
      "epoch": 0.7433628318584071,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0007933901068425538,
      "loss": 0.8961,
      "mean_token_accuracy": 0.7628611207008362,
      "step": 105
    },
    {
      "epoch": 0.7787610619469026,
      "grad_norm": 0.5859375,
      "learning_rate": 0.000767703677935813,
      "loss": 0.891,
      "mean_token_accuracy": 0.7617514222860337,
      "step": 110
    },
    {
      "epoch": 0.8141592920353983,
      "grad_norm": 0.69140625,
      "learning_rate": 0.000740985643796569,
      "loss": 0.9057,
      "mean_token_accuracy": 0.7621735870838166,
      "step": 115
    },
    {
      "epoch": 0.8495575221238938,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0007133389632785542,
      "loss": 0.8874,
      "mean_token_accuracy": 0.7660593271255494,
      "step": 120
    },
    {
      "epoch": 0.8849557522123894,
      "grad_norm": 0.5703125,
      "learning_rate": 0.0006848701738059226,
      "loss": 0.8441,
      "mean_token_accuracy": 0.7740032315254212,
      "step": 125
    },
    {
      "epoch": 0.9203539823008849,
      "grad_norm": 0.59375,
      "learning_rate": 0.0006556889808276594,
      "loss": 0.8767,
      "mean_token_accuracy": 0.7661084860563279,
      "step": 130
    },
    {
      "epoch": 0.9557522123893806,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0006259078350639009,
      "loss": 0.8962,
      "mean_token_accuracy": 0.7660992830991745,
      "step": 135
    },
    {
      "epoch": 0.9911504424778761,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0005956414991732583,
      "loss": 0.8515,
      "mean_token_accuracy": 0.7711150646209717,
      "step": 140
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.8502547144889832,
      "eval_mean_token_accuracy": 0.7765891637120929,
      "eval_runtime": 2.1657,
      "eval_samples_per_second": 54.949,
      "eval_steps_per_second": 13.853,
      "step": 142
    },
    {
      "epoch": 0.928,
      "grad_norm": 1.140625,
      "learning_rate": 0.000649163469284578,
      "loss": 1.5122,
      "mean_token_accuracy": 0.6797654032707214,
      "step": 145
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.6484375,
      "learning_rate": 0.000622170203068947,
      "loss": 1.1682,
      "mean_token_accuracy": 0.7231339365243912,
      "step": 150
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0005947925441958392,
      "loss": 1.1885,
      "mean_token_accuracy": 0.7220241338014602,
      "step": 155
    },
    {
      "epoch": 0.9984,
      "eval_loss": 1.164101481437683,
      "eval_mean_token_accuracy": 0.7249869391029956,
      "eval_runtime": 67.8529,
      "eval_samples_per_second": 14.738,
      "eval_steps_per_second": 3.684,
      "step": 156
    },
    {
      "epoch": 1.0256,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0005671166329088278,
      "loss": 1.3738,
      "mean_token_accuracy": 0.7169055119156837,
      "step": 160
    },
    {
      "epoch": 1.0576,
      "grad_norm": 0.578125,
      "learning_rate": 0.0005392295478639225,
      "loss": 1.0767,
      "mean_token_accuracy": 0.7402106642723083,
      "step": 165
    },
    {
      "epoch": 1.0896,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0005112190321479025,
      "loss": 1.0308,
      "mean_token_accuracy": 0.7459876298904419,
      "step": 170
    },
    {
      "epoch": 1.1216,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0004831732172061032,
      "loss": 1.1519,
      "mean_token_accuracy": 0.7256286978721619,
      "step": 175
    },
    {
      "epoch": 1.1536,
      "grad_norm": 0.46484375,
      "learning_rate": 0.0004551803455482833,
      "loss": 1.0695,
      "mean_token_accuracy": 0.739355456829071,
      "step": 180
    },
    {
      "epoch": 1.1856,
      "grad_norm": 0.369140625,
      "learning_rate": 0.0004273284931050438,
      "loss": 1.0264,
      "mean_token_accuracy": 0.7484620213508606,
      "step": 185
    },
    {
      "epoch": 1.2176,
      "grad_norm": 0.439453125,
      "learning_rate": 0.00039970529210836363,
      "loss": 1.1083,
      "mean_token_accuracy": 0.7330797225236892,
      "step": 190
    },
    {
      "epoch": 1.2496,
      "grad_norm": 0.431640625,
      "learning_rate": 0.00037239765536817873,
      "loss": 0.9954,
      "mean_token_accuracy": 0.7526825368404388,
      "step": 195
    },
    {
      "epoch": 1.2816,
      "grad_norm": 0.369140625,
      "learning_rate": 0.00034549150281252633,
      "loss": 0.9803,
      "mean_token_accuracy": 0.754820802807808,
      "step": 200
    },
    {
      "epoch": 1.3136,
      "grad_norm": 0.357421875,
      "learning_rate": 0.000319071491151664,
      "loss": 0.9398,
      "mean_token_accuracy": 0.762332808971405,
      "step": 205
    },
    {
      "epoch": 1.3456000000000001,
      "grad_norm": 0.40625,
      "learning_rate": 0.00029322074751673977,
      "loss": 0.9961,
      "mean_token_accuracy": 0.7532915711402893,
      "step": 210
    },
    {
      "epoch": 1.3776,
      "grad_norm": 0.380859375,
      "learning_rate": 0.000268020607911083,
      "loss": 0.9954,
      "mean_token_accuracy": 0.7548110991716385,
      "step": 215
    },
    {
      "epoch": 1.4096,
      "grad_norm": 0.349609375,
      "learning_rate": 0.000243550361297047,
      "loss": 0.975,
      "mean_token_accuracy": 0.7560653567314148,
      "step": 220
    },
    {
      "epoch": 1.4416,
      "grad_norm": 0.412109375,
      "learning_rate": 0.00021988700012359863,
      "loss": 1.0117,
      "mean_token_accuracy": 0.7497000277042389,
      "step": 225
    },
    {
      "epoch": 1.4736,
      "grad_norm": 0.357421875,
      "learning_rate": 0.0001971049780795901,
      "loss": 0.9645,
      "mean_token_accuracy": 0.7578672379255295,
      "step": 230
    },
    {
      "epoch": 1.5056,
      "grad_norm": 0.37890625,
      "learning_rate": 0.00017527597583490823,
      "loss": 0.9122,
      "mean_token_accuracy": 0.7679250746965408,
      "step": 235
    },
    {
      "epoch": 1.5375999999999999,
      "grad_norm": 0.38671875,
      "learning_rate": 0.00015446867550656773,
      "loss": 1.0358,
      "mean_token_accuracy": 0.7469679594039917,
      "step": 240
    },
    {
      "epoch": 1.5695999999999999,
      "grad_norm": 0.400390625,
      "learning_rate": 0.0001347485445593612,
      "loss": 0.9479,
      "mean_token_accuracy": 0.7598189502954483,
      "step": 245
    },
    {
      "epoch": 1.6016,
      "grad_norm": 0.34765625,
      "learning_rate": 0.00011617762982099444,
      "loss": 0.9622,
      "mean_token_accuracy": 0.7569534540176391,
      "step": 250
    },
    {
      "epoch": 1.6336,
      "grad_norm": 0.4140625,
      "learning_rate": 9.881436225981105e-05,
      "loss": 0.9807,
      "mean_token_accuracy": 0.7524196356534958,
      "step": 255
    },
    {
      "epoch": 1.6656,
      "grad_norm": 0.435546875,
      "learning_rate": 8.271337313934873e-05,
      "loss": 0.9374,
      "mean_token_accuracy": 0.765364584326744,
      "step": 260
    },
    {
      "epoch": 1.6976,
      "grad_norm": 0.408203125,
      "learning_rate": 6.792532212817271e-05,
      "loss": 0.9754,
      "mean_token_accuracy": 0.7565860599279404,
      "step": 265
    },
    {
      "epoch": 1.7296,
      "grad_norm": 0.44921875,
      "learning_rate": 5.449673790581611e-05,
      "loss": 0.9346,
      "mean_token_accuracy": 0.7654347807168961,
      "step": 270
    },
    {
      "epoch": 1.7616,
      "grad_norm": 0.44140625,
      "learning_rate": 4.2469871766340095e-05,
      "loss": 1.0262,
      "mean_token_accuracy": 0.7484832376241684,
      "step": 275
    },
    {
      "epoch": 1.7936,
      "grad_norm": 0.41796875,
      "learning_rate": 3.18825646801314e-05,
      "loss": 0.9733,
      "mean_token_accuracy": 0.7553656280040741,
      "step": 280
    },
    {
      "epoch": 1.8256000000000001,
      "grad_norm": 0.36328125,
      "learning_rate": 2.276812823220964e-05,
      "loss": 0.948,
      "mean_token_accuracy": 0.7614956557750702,
      "step": 285
    },
    {
      "epoch": 1.8576000000000001,
      "grad_norm": 0.392578125,
      "learning_rate": 1.5155239811656562e-05,
      "loss": 0.9761,
      "mean_token_accuracy": 0.753793916106224,
      "step": 290
    },
    {
      "epoch": 1.8896,
      "grad_norm": 0.41015625,
      "learning_rate": 9.0678523819408e-06,
      "loss": 0.9292,
      "mean_token_accuracy": 0.763578137755394,
      "step": 295
    },
    {
      "epoch": 1.9216,
      "grad_norm": 0.416015625,
      "learning_rate": 4.52511911603265e-06,
      "loss": 0.9288,
      "mean_token_accuracy": 0.7666039824485779,
      "step": 300
    },
    {
      "epoch": 1.9536,
      "grad_norm": 0.357421875,
      "learning_rate": 1.541333133436018e-06,
      "loss": 0.995,
      "mean_token_accuracy": 0.7530867516994476,
      "step": 305
    },
    {
      "epoch": 1.9856,
      "grad_norm": 0.388671875,
      "learning_rate": 1.2588252874673466e-07,
      "loss": 0.9187,
      "mean_token_accuracy": 0.7661501169204712,
      "step": 310
    },
    {
      "epoch": 1.9984,
      "eval_loss": 1.0250508785247803,
      "eval_mean_token_accuracy": 0.7480007416980211,
      "eval_runtime": 70.0256,
      "eval_samples_per_second": 14.28,
      "eval_steps_per_second": 3.57,
      "step": 312
    },
    {
      "epoch": 1.9984,
      "step": 312,
      "total_flos": 6258600061401600.0,
      "train_loss": 0.5569620216504122,
      "train_runtime": 888.3293,
      "train_samples_per_second": 5.629,
      "train_steps_per_second": 0.351
    }
  ],
  "logging_steps": 5,
  "max_steps": 312,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6258600061401600.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}