{
  "best_metric": 1.362037181854248,
  "best_model_checkpoint": "data/Gemma-2-2B_task-1_60-samples_config-1/checkpoint-34",
  "epoch": 12.869565217391305,
  "eval_steps": 500,
  "global_step": 74,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17391304347826086,
      "grad_norm": 2.9326045513153076,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.4121,
      "step": 1
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 2.785604238510132,
      "learning_rate": 8.000000000000001e-06,
      "loss": 2.5463,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 2.592984914779663,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 2.489,
      "step": 4
    },
    {
      "epoch": 0.8695652173913043,
      "eval_loss": 2.2686474323272705,
      "eval_runtime": 1.0904,
      "eval_samples_per_second": 11.005,
      "eval_steps_per_second": 11.005,
      "step": 5
    },
    {
      "epoch": 1.0434782608695652,
      "grad_norm": 2.600557804107666,
      "learning_rate": 2.4e-05,
      "loss": 2.3738,
      "step": 6
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 2.8645849227905273,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 2.3608,
      "step": 8
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 2.449917793273926,
      "learning_rate": 4e-05,
      "loss": 2.1234,
      "step": 10
    },
    {
      "epoch": 1.9130434782608696,
      "eval_loss": 1.8447164297103882,
      "eval_runtime": 1.0662,
      "eval_samples_per_second": 11.255,
      "eval_steps_per_second": 11.255,
      "step": 11
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 2.242774486541748,
      "learning_rate": 4.8e-05,
      "loss": 1.9407,
      "step": 12
    },
    {
      "epoch": 2.4347826086956523,
      "grad_norm": 1.8629240989685059,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 1.7363,
      "step": 14
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 1.4939218759536743,
      "learning_rate": 6.400000000000001e-05,
      "loss": 1.5886,
      "step": 16
    },
    {
      "epoch": 2.9565217391304346,
      "eval_loss": 1.5018290281295776,
      "eval_runtime": 1.2074,
      "eval_samples_per_second": 9.939,
      "eval_steps_per_second": 9.939,
      "step": 17
    },
    {
      "epoch": 3.130434782608696,
      "grad_norm": 1.2370601892471313,
      "learning_rate": 7.2e-05,
      "loss": 1.471,
      "step": 18
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 1.2678393125534058,
      "learning_rate": 8e-05,
      "loss": 1.3431,
      "step": 20
    },
    {
      "epoch": 3.8260869565217392,
      "grad_norm": 1.1590300798416138,
      "learning_rate": 8.800000000000001e-05,
      "loss": 1.4359,
      "step": 22
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.3991178274154663,
      "eval_runtime": 1.196,
      "eval_samples_per_second": 10.033,
      "eval_steps_per_second": 10.033,
      "step": 23
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 1.0870388746261597,
      "learning_rate": 9.6e-05,
      "loss": 1.2965,
      "step": 24
    },
    {
      "epoch": 4.521739130434782,
      "grad_norm": 0.9513787031173706,
      "learning_rate": 9.999512620046522e-05,
      "loss": 1.216,
      "step": 26
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.9963628649711609,
      "learning_rate": 9.995614150494293e-05,
      "loss": 1.1595,
      "step": 28
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 1.3781918287277222,
      "eval_runtime": 1.2257,
      "eval_samples_per_second": 9.79,
      "eval_steps_per_second": 9.79,
      "step": 28
    },
    {
      "epoch": 5.217391304347826,
      "grad_norm": 1.009848952293396,
      "learning_rate": 9.987820251299122e-05,
      "loss": 1.1039,
      "step": 30
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 1.2583975791931152,
      "learning_rate": 9.976136999909156e-05,
      "loss": 1.0074,
      "step": 32
    },
    {
      "epoch": 5.913043478260869,
      "grad_norm": 1.1578857898712158,
      "learning_rate": 9.96057350657239e-05,
      "loss": 0.9043,
      "step": 34
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 1.362037181854248,
      "eval_runtime": 1.0954,
      "eval_samples_per_second": 10.955,
      "eval_steps_per_second": 10.955,
      "step": 34
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 1.0541073083877563,
      "learning_rate": 9.941141907232765e-05,
      "loss": 0.7793,
      "step": 36
    },
    {
      "epoch": 6.608695652173913,
      "grad_norm": 1.4098272323608398,
      "learning_rate": 9.917857354066931e-05,
      "loss": 0.7546,
      "step": 38
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 1.4439623355865479,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.773,
      "step": 40
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 1.454060673713684,
      "eval_runtime": 1.2372,
      "eval_samples_per_second": 9.699,
      "eval_steps_per_second": 9.699,
      "step": 40
    },
    {
      "epoch": 7.304347826086957,
      "grad_norm": 1.3048129081726074,
      "learning_rate": 9.859805002892732e-05,
      "loss": 0.6603,
      "step": 42
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 1.3808172941207886,
      "learning_rate": 9.825082472361557e-05,
      "loss": 0.5411,
      "step": 44
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.6105766296386719,
      "learning_rate": 9.786597487660337e-05,
      "loss": 0.4933,
      "step": 46
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.616339087486267,
      "eval_runtime": 1.0573,
      "eval_samples_per_second": 11.349,
      "eval_steps_per_second": 11.349,
      "step": 46
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 1.46830153465271,
      "learning_rate": 9.744380058222483e-05,
      "loss": 0.3801,
      "step": 48
    },
    {
      "epoch": 8.695652173913043,
      "grad_norm": 1.603589653968811,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.354,
      "step": 50
    },
    {
      "epoch": 8.869565217391305,
      "eval_loss": 1.9201489686965942,
      "eval_runtime": 1.0601,
      "eval_samples_per_second": 11.32,
      "eval_steps_per_second": 11.32,
      "step": 51
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 1.9514093399047852,
      "learning_rate": 9.648882429441257e-05,
      "loss": 0.3032,
      "step": 52
    },
    {
      "epoch": 9.391304347826088,
      "grad_norm": 1.9017702341079712,
      "learning_rate": 9.595676696276172e-05,
      "loss": 0.2081,
      "step": 54
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 1.7318685054779053,
      "learning_rate": 9.538887392664544e-05,
      "loss": 0.1478,
      "step": 56
    },
    {
      "epoch": 9.91304347826087,
      "eval_loss": 2.480901002883911,
      "eval_runtime": 1.1179,
      "eval_samples_per_second": 10.734,
      "eval_steps_per_second": 10.734,
      "step": 57
    },
    {
      "epoch": 10.08695652173913,
      "grad_norm": 1.9642924070358276,
      "learning_rate": 9.478558801197065e-05,
      "loss": 0.1671,
      "step": 58
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 1.2309162616729736,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0836,
      "step": 60
    },
    {
      "epoch": 10.782608695652174,
      "grad_norm": 1.401197910308838,
      "learning_rate": 9.347474647526095e-05,
      "loss": 0.0676,
      "step": 62
    },
    {
      "epoch": 10.956521739130435,
      "eval_loss": 2.9183950424194336,
      "eval_runtime": 1.0831,
      "eval_samples_per_second": 11.079,
      "eval_steps_per_second": 11.079,
      "step": 63
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 1.3135957717895508,
      "learning_rate": 9.276821300802534e-05,
      "loss": 0.0935,
      "step": 64
    },
    {
      "epoch": 11.478260869565217,
      "grad_norm": 1.3647031784057617,
      "learning_rate": 9.202833017478422e-05,
      "loss": 0.0463,
      "step": 66
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.8959400653839111,
      "learning_rate": 9.125567491391476e-05,
      "loss": 0.0308,
      "step": 68
    },
    {
      "epoch": 12.0,
      "eval_loss": 3.2972118854522705,
      "eval_runtime": 1.2332,
      "eval_samples_per_second": 9.731,
      "eval_steps_per_second": 9.731,
      "step": 69
    },
    {
      "epoch": 12.173913043478262,
      "grad_norm": 2.00899338722229,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0286,
      "step": 70
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 2.88161563873291,
      "learning_rate": 8.961448216775954e-05,
      "loss": 0.0286,
      "step": 72
    },
    {
      "epoch": 12.869565217391305,
      "grad_norm": 1.5414081811904907,
      "learning_rate": 8.874722443520899e-05,
      "loss": 0.0342,
      "step": 74
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 3.367413282394409,
      "eval_runtime": 1.1245,
      "eval_samples_per_second": 10.672,
      "eval_steps_per_second": 10.672,
      "step": 74
    },
    {
      "epoch": 12.869565217391305,
      "step": 74,
      "total_flos": 2319598189281280.0,
      "train_loss": 0.9195622371862063,
      "train_runtime": 286.9716,
      "train_samples_per_second": 8.015,
      "train_steps_per_second": 0.871
    }
  ],
  "logging_steps": 2,
  "max_steps": 250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 25,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 7,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2319598189281280.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}