{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.3102506825515016,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0062050136510300325,
      "grad_norm": 101.88961029052734,
      "learning_rate": 0.0001,
      "loss": 4.5088,
      "step": 100
    },
    {
      "epoch": 0.012410027302060065,
      "grad_norm": 99.7363510131836,
      "learning_rate": 9.999038127056248e-05,
      "loss": 0.8644,
      "step": 200
    },
    {
      "epoch": 0.018615040953090096,
      "grad_norm": 59.41786575317383,
      "learning_rate": 9.996152878304816e-05,
      "loss": 0.7189,
      "step": 300
    },
    {
      "epoch": 0.02482005460412013,
      "grad_norm": 61.05416488647461,
      "learning_rate": 9.991345363842789e-05,
      "loss": 0.6253,
      "step": 400
    },
    {
      "epoch": 0.03102506825515016,
      "grad_norm": 53.87808609008789,
      "learning_rate": 9.9846174333574e-05,
      "loss": 0.5863,
      "step": 500
    },
    {
      "epoch": 0.03723008190618019,
      "grad_norm": 48.58293914794922,
      "learning_rate": 9.975971675414371e-05,
      "loss": 0.5555,
      "step": 600
    },
    {
      "epoch": 0.04343509555721023,
      "grad_norm": 42.353153228759766,
      "learning_rate": 9.965411416461959e-05,
      "loss": 0.5566,
      "step": 700
    },
    {
      "epoch": 0.04964010920824026,
      "grad_norm": 46.06748580932617,
      "learning_rate": 9.952940719551112e-05,
      "loss": 0.5425,
      "step": 800
    },
    {
      "epoch": 0.05584512285927029,
      "grad_norm": 46.486637115478516,
      "learning_rate": 9.938564382772205e-05,
      "loss": 0.5004,
      "step": 900
    },
    {
      "epoch": 0.06205013651030032,
      "grad_norm": 43.99421691894531,
      "learning_rate": 9.922287937408994e-05,
      "loss": 0.4722,
      "step": 1000
    },
    {
      "epoch": 0.06205013651030032,
      "eval_avg_non_pair_similarity": -0.00011276310920446573,
      "eval_avg_pair_similarity": 0.0020758428336121143,
      "eval_loss": 0.44835320115089417,
      "eval_runtime": 52.6112,
      "eval_samples_per_second": 9.504,
      "eval_similarity_ratio": -18.408882552609725,
      "eval_steps_per_second": 0.304,
      "step": 1000
    },
    {
      "epoch": 0.06825515016133035,
      "grad_norm": 38.14698791503906,
      "learning_rate": 9.904117645810441e-05,
      "loss": 0.4444,
      "step": 1100
    },
    {
      "epoch": 0.07446016381236038,
      "grad_norm": 29.508113861083984,
      "learning_rate": 9.884060498981296e-05,
      "loss": 0.4615,
      "step": 1200
    },
    {
      "epoch": 0.08066517746339041,
      "grad_norm": 34.64622497558594,
      "learning_rate": 9.862124213892304e-05,
      "loss": 0.4665,
      "step": 1300
    },
    {
      "epoch": 0.08687019111442046,
      "grad_norm": 40.75010299682617,
      "learning_rate": 9.838317230511112e-05,
      "loss": 0.4275,
      "step": 1400
    },
    {
      "epoch": 0.09307520476545049,
      "grad_norm": 28.284589767456055,
      "learning_rate": 9.81264870855499e-05,
      "loss": 0.4129,
      "step": 1500
    },
    {
      "epoch": 0.09928021841648052,
      "grad_norm": 34.2197151184082,
      "learning_rate": 9.785128523966653e-05,
      "loss": 0.4389,
      "step": 1600
    },
    {
      "epoch": 0.10548523206751055,
      "grad_norm": 27.808895111083984,
      "learning_rate": 9.755767265114484e-05,
      "loss": 0.4184,
      "step": 1700
    },
    {
      "epoch": 0.11169024571854058,
      "grad_norm": 35.00907897949219,
      "learning_rate": 9.724576228718678e-05,
      "loss": 0.3995,
      "step": 1800
    },
    {
      "epoch": 0.11789525936957061,
      "grad_norm": 30.57769203186035,
      "learning_rate": 9.691567415504832e-05,
      "loss": 0.415,
      "step": 1900
    },
    {
      "epoch": 0.12410027302060064,
      "grad_norm": 26.989404678344727,
      "learning_rate": 9.656753525586681e-05,
      "loss": 0.4052,
      "step": 2000
    },
    {
      "epoch": 0.12410027302060064,
      "eval_avg_non_pair_similarity": 0.0006865022985091086,
      "eval_avg_pair_similarity": 0.007665629971772433,
      "eval_loss": 0.3601702153682709,
      "eval_runtime": 52.565,
      "eval_samples_per_second": 9.512,
      "eval_similarity_ratio": 11.166211662247951,
      "eval_steps_per_second": 0.304,
      "step": 2000
    },
    {
      "epoch": 0.1303052866716307,
      "grad_norm": 32.37137222290039,
      "learning_rate": 9.620147953579737e-05,
      "loss": 0.3979,
      "step": 2100
    },
    {
      "epoch": 0.1365103003226607,
      "grad_norm": 22.6475772857666,
      "learning_rate": 9.581764783447719e-05,
      "loss": 0.3807,
      "step": 2200
    },
    {
      "epoch": 0.14271531397369075,
      "grad_norm": 22.612077713012695,
      "learning_rate": 9.54161878308377e-05,
      "loss": 0.3711,
      "step": 2300
    },
    {
      "epoch": 0.14892032762472077,
      "grad_norm": 30.973310470581055,
      "learning_rate": 9.499725398628507e-05,
      "loss": 0.3658,
      "step": 2400
    },
    {
      "epoch": 0.1551253412757508,
      "grad_norm": 24.443492889404297,
      "learning_rate": 9.456100748527143e-05,
      "loss": 0.37,
      "step": 2500
    },
    {
      "epoch": 0.16133035492678083,
      "grad_norm": 33.75666809082031,
      "learning_rate": 9.410761617327921e-05,
      "loss": 0.3512,
      "step": 2600
    },
    {
      "epoch": 0.16753536857781087,
      "grad_norm": 23.727365493774414,
      "learning_rate": 9.363725449224282e-05,
      "loss": 0.3318,
      "step": 2700
    },
    {
      "epoch": 0.17374038222884092,
      "grad_norm": 28.48720359802246,
      "learning_rate": 9.315010341343213e-05,
      "loss": 0.3729,
      "step": 2800
    },
    {
      "epoch": 0.17994539587987093,
      "grad_norm": 28.53714370727539,
      "learning_rate": 9.264635036782405e-05,
      "loss": 0.3369,
      "step": 2900
    },
    {
      "epoch": 0.18615040953090098,
      "grad_norm": 21.766740798950195,
      "learning_rate": 9.212618917398855e-05,
      "loss": 0.317,
      "step": 3000
    },
    {
      "epoch": 0.18615040953090098,
      "eval_avg_non_pair_similarity": 0.00025785160030682527,
      "eval_avg_pair_similarity": 0.0073178326906636355,
      "eval_loss": 0.323689728975296,
      "eval_runtime": 52.8381,
      "eval_samples_per_second": 9.463,
      "eval_similarity_ratio": 28.38001657525464,
      "eval_steps_per_second": 0.303,
      "step": 3000
    },
    {
      "epoch": 0.192355423181931,
      "grad_norm": 25.073217391967773,
      "learning_rate": 9.158981996351711e-05,
      "loss": 0.3092,
      "step": 3100
    },
    {
      "epoch": 0.19856043683296104,
      "grad_norm": 25.02987289428711,
      "learning_rate": 9.10374491040223e-05,
      "loss": 0.3177,
      "step": 3200
    },
    {
      "epoch": 0.20476545048399106,
      "grad_norm": 23.147781372070312,
      "learning_rate": 9.046928911973799e-05,
      "loss": 0.3357,
      "step": 3300
    },
    {
      "epoch": 0.2109704641350211,
      "grad_norm": 18.975126266479492,
      "learning_rate": 8.988555860975082e-05,
      "loss": 0.2946,
      "step": 3400
    },
    {
      "epoch": 0.21717547778605112,
      "grad_norm": 22.149898529052734,
      "learning_rate": 8.92864821638944e-05,
      "loss": 0.3184,
      "step": 3500
    },
    {
      "epoch": 0.22338049143708116,
      "grad_norm": 31.714447021484375,
      "learning_rate": 8.86722902763385e-05,
      "loss": 0.315,
      "step": 3600
    },
    {
      "epoch": 0.22958550508811118,
      "grad_norm": 22.137062072753906,
      "learning_rate": 8.804321925690672e-05,
      "loss": 0.3007,
      "step": 3700
    },
    {
      "epoch": 0.23579051873914122,
      "grad_norm": 27.34349250793457,
      "learning_rate": 8.739951114015636e-05,
      "loss": 0.3297,
      "step": 3800
    },
    {
      "epoch": 0.24199553239017127,
      "grad_norm": 20.715852737426758,
      "learning_rate": 8.674141359225589e-05,
      "loss": 0.2943,
      "step": 3900
    },
    {
      "epoch": 0.2482005460412013,
      "grad_norm": 21.26046371459961,
      "learning_rate": 8.60691798156956e-05,
      "loss": 0.284,
      "step": 4000
    },
    {
      "epoch": 0.2482005460412013,
      "eval_avg_non_pair_similarity": 0.00036512027781798257,
      "eval_avg_pair_similarity": 0.002839699084404856,
      "eval_loss": 0.26940277218818665,
      "eval_runtime": 52.6691,
      "eval_samples_per_second": 9.493,
      "eval_similarity_ratio": 7.777434606961175,
      "eval_steps_per_second": 0.304,
      "step": 4000
    },
    {
      "epoch": 0.25440555969223133,
      "grad_norm": 22.599706649780273,
      "learning_rate": 8.538306845186808e-05,
      "loss": 0.2942,
      "step": 4100
    },
    {
      "epoch": 0.2606105733432614,
      "grad_norm": 16.177858352661133,
      "learning_rate": 8.468334348155624e-05,
      "loss": 0.2874,
      "step": 4200
    },
    {
      "epoch": 0.26681558699429136,
      "grad_norm": 15.479412078857422,
      "learning_rate": 8.39702741233669e-05,
      "loss": 0.2758,
      "step": 4300
    },
    {
      "epoch": 0.2730206006453214,
      "grad_norm": 20.40460777282715,
      "learning_rate": 8.324413473014913e-05,
      "loss": 0.2788,
      "step": 4400
    },
    {
      "epoch": 0.27922561429635145,
      "grad_norm": 15.524222373962402,
      "learning_rate": 8.250520468343722e-05,
      "loss": 0.2667,
      "step": 4500
    },
    {
      "epoch": 0.2854306279473815,
      "grad_norm": 19.124486923217773,
      "learning_rate": 8.175376828595885e-05,
      "loss": 0.2828,
      "step": 4600
    },
    {
      "epoch": 0.29163564159841154,
      "grad_norm": 14.729342460632324,
      "learning_rate": 8.09901146522499e-05,
      "loss": 0.2757,
      "step": 4700
    },
    {
      "epoch": 0.29784065524944153,
      "grad_norm": 19.547435760498047,
      "learning_rate": 8.021453759741784e-05,
      "loss": 0.3102,
      "step": 4800
    },
    {
      "epoch": 0.3040456689004716,
      "grad_norm": 16.095699310302734,
      "learning_rate": 7.942733552409662e-05,
      "loss": 0.2791,
      "step": 4900
    },
    {
      "epoch": 0.3102506825515016,
      "grad_norm": 25.538352966308594,
      "learning_rate": 7.862881130763646e-05,
      "loss": 0.2487,
      "step": 5000
    },
    {
      "epoch": 0.3102506825515016,
      "eval_avg_non_pair_similarity": -2.6742616108896766e-05,
      "eval_avg_pair_similarity": 0.007982193220406771,
      "eval_loss": 0.3381979465484619,
      "eval_runtime": 52.9305,
      "eval_samples_per_second": 9.446,
      "eval_similarity_ratio": -298.4821375703496,
      "eval_steps_per_second": 0.302,
      "step": 5000
    }
  ],
  "logging_steps": 100,
  "max_steps": 16116,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}