{
  "best_metric": 3.1670639514923096,
  "best_model_checkpoint": "./output/models/gpt2-medium-wikitext/checkpoint-8500",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 8910,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05611672278338945,
      "grad_norm": 1.7955036163330078,
      "learning_rate": 1.1223344556677892e-05,
      "loss": 8.9349,
      "step": 100
    },
    {
      "epoch": 0.1122334455667789,
      "grad_norm": 1.2832236289978027,
      "learning_rate": 2.2446689113355783e-05,
      "loss": 7.3426,
      "step": 200
    },
    {
      "epoch": 0.16835016835016836,
      "grad_norm": 1.585518717765808,
      "learning_rate": 3.3670033670033675e-05,
      "loss": 6.6306,
      "step": 300
    },
    {
      "epoch": 0.2244668911335578,
      "grad_norm": 1.6020684242248535,
      "learning_rate": 4.4893378226711566e-05,
      "loss": 6.3121,
      "step": 400
    },
    {
      "epoch": 0.28058361391694725,
      "grad_norm": 1.7420004606246948,
      "learning_rate": 5.611672278338945e-05,
      "loss": 6.0809,
      "step": 500
    },
    {
      "epoch": 0.28058361391694725,
      "eval_accuracy": 0.1883480988962881,
      "eval_bleu": 0.03330409357604442,
      "eval_loss": 5.957973480224609,
      "eval_perplexity": 386.8254200337184,
      "eval_runtime": 19.389,
      "eval_samples_per_second": 58.848,
      "eval_steps_per_second": 0.928,
      "step": 500
    },
    {
      "epoch": 0.3367003367003367,
      "grad_norm": 1.2030788660049438,
      "learning_rate": 6.734006734006735e-05,
      "loss": 5.8698,
      "step": 600
    },
    {
      "epoch": 0.39281705948372614,
      "grad_norm": 1.6605628728866577,
      "learning_rate": 7.856341189674523e-05,
      "loss": 5.6555,
      "step": 700
    },
    {
      "epoch": 0.4489337822671156,
      "grad_norm": 1.6018306016921997,
      "learning_rate": 8.978675645342313e-05,
      "loss": 5.4756,
      "step": 800
    },
    {
      "epoch": 0.5050505050505051,
      "grad_norm": 1.2040090560913086,
      "learning_rate": 9.988776655443322e-05,
      "loss": 5.2646,
      "step": 900
    },
    {
      "epoch": 0.5611672278338945,
      "grad_norm": 1.292252779006958,
      "learning_rate": 9.864072827035791e-05,
      "loss": 5.0644,
      "step": 1000
    },
    {
      "epoch": 0.5611672278338945,
      "eval_accuracy": 0.26234126056014045,
      "eval_bleu": 0.0650543285360525,
      "eval_loss": 4.919076442718506,
      "eval_perplexity": 136.87614183623467,
      "eval_runtime": 14.9502,
      "eval_samples_per_second": 76.32,
      "eval_steps_per_second": 1.204,
      "step": 1000
    },
    {
      "epoch": 0.6172839506172839,
      "grad_norm": 0.9987787008285522,
      "learning_rate": 9.73936899862826e-05,
      "loss": 4.8884,
      "step": 1100
    },
    {
      "epoch": 0.6734006734006734,
      "grad_norm": 0.983322262763977,
      "learning_rate": 9.614665170220725e-05,
      "loss": 4.7342,
      "step": 1200
    },
    {
      "epoch": 0.7295173961840629,
      "grad_norm": 0.9396106004714966,
      "learning_rate": 9.489961341813194e-05,
      "loss": 4.5934,
      "step": 1300
    },
    {
      "epoch": 0.7856341189674523,
      "grad_norm": 0.9439830780029297,
      "learning_rate": 9.365257513405662e-05,
      "loss": 4.4643,
      "step": 1400
    },
    {
      "epoch": 0.8417508417508418,
      "grad_norm": 0.9339156150817871,
      "learning_rate": 9.24055368499813e-05,
      "loss": 4.3331,
      "step": 1500
    },
    {
      "epoch": 0.8417508417508418,
      "eval_accuracy": 0.32261577066643365,
      "eval_bleu": 0.0889997104198564,
      "eval_loss": 4.212368965148926,
      "eval_perplexity": 67.51629425712589,
      "eval_runtime": 19.4999,
      "eval_samples_per_second": 58.513,
      "eval_steps_per_second": 0.923,
      "step": 1500
    },
    {
      "epoch": 0.8978675645342312,
      "grad_norm": 0.8961524367332458,
      "learning_rate": 9.115849856590598e-05,
      "loss": 4.2464,
      "step": 1600
    },
    {
      "epoch": 0.9539842873176206,
      "grad_norm": 0.844565749168396,
      "learning_rate": 8.991146028183066e-05,
      "loss": 4.1434,
      "step": 1700
    },
    {
      "epoch": 1.0101010101010102,
      "grad_norm": 0.8904160857200623,
      "learning_rate": 8.866442199775533e-05,
      "loss": 4.0797,
      "step": 1800
    },
    {
      "epoch": 1.0662177328843996,
      "grad_norm": 0.8723273277282715,
      "learning_rate": 8.741738371368002e-05,
      "loss": 3.9988,
      "step": 1900
    },
    {
      "epoch": 1.122334455667789,
      "grad_norm": 0.8569739460945129,
      "learning_rate": 8.617034542960469e-05,
      "loss": 3.9451,
      "step": 2000
    },
    {
      "epoch": 1.122334455667789,
      "eval_accuracy": 0.35319123781423406,
      "eval_bleu": 0.1090324206065986,
      "eval_loss": 3.8835041522979736,
      "eval_perplexity": 48.59419854591979,
      "eval_runtime": 19.5147,
      "eval_samples_per_second": 58.469,
      "eval_steps_per_second": 0.922,
      "step": 2000
    },
    {
      "epoch": 1.1784511784511784,
      "grad_norm": 0.8484457731246948,
      "learning_rate": 8.492330714552937e-05,
      "loss": 3.9031,
      "step": 2100
    },
    {
      "epoch": 1.2345679012345678,
      "grad_norm": 0.8243273496627808,
      "learning_rate": 8.367626886145406e-05,
      "loss": 3.856,
      "step": 2200
    },
    {
      "epoch": 1.2906846240179575,
      "grad_norm": 0.7708187103271484,
      "learning_rate": 8.242923057737873e-05,
      "loss": 3.8365,
      "step": 2300
    },
    {
      "epoch": 1.3468013468013469,
      "grad_norm": 0.8010033369064331,
      "learning_rate": 8.11821922933034e-05,
      "loss": 3.7951,
      "step": 2400
    },
    {
      "epoch": 1.4029180695847363,
      "grad_norm": 0.7792832851409912,
      "learning_rate": 7.993515400922809e-05,
      "loss": 3.7568,
      "step": 2500
    },
    {
      "epoch": 1.4029180695847363,
      "eval_accuracy": 0.36843313688752044,
      "eval_bleu": 0.12256078544379752,
      "eval_loss": 3.7051432132720947,
      "eval_perplexity": 40.655869360052804,
      "eval_runtime": 19.518,
      "eval_samples_per_second": 58.459,
      "eval_steps_per_second": 0.922,
      "step": 2500
    },
    {
      "epoch": 1.4590347923681257,
      "grad_norm": 0.7673516273498535,
      "learning_rate": 7.868811572515277e-05,
      "loss": 3.7381,
      "step": 2600
    },
    {
      "epoch": 1.5151515151515151,
      "grad_norm": 0.756196916103363,
      "learning_rate": 7.744107744107744e-05,
      "loss": 3.7082,
      "step": 2700
    },
    {
      "epoch": 1.5712682379349046,
      "grad_norm": 0.770793080329895,
      "learning_rate": 7.619403915700213e-05,
      "loss": 3.6751,
      "step": 2800
    },
    {
      "epoch": 1.627384960718294,
      "grad_norm": 0.7664377689361572,
      "learning_rate": 7.49470008729268e-05,
      "loss": 3.6617,
      "step": 2900
    },
    {
      "epoch": 1.6835016835016834,
      "grad_norm": 0.8182048201560974,
      "learning_rate": 7.369996258885148e-05,
      "loss": 3.6478,
      "step": 3000
    },
    {
      "epoch": 1.6835016835016834,
      "eval_accuracy": 0.37868464407154295,
      "eval_bleu": 0.13106863900596108,
      "eval_loss": 3.5827126502990723,
      "eval_perplexity": 35.970985325012926,
      "eval_runtime": 19.6802,
      "eval_samples_per_second": 57.977,
      "eval_steps_per_second": 0.915,
      "step": 3000
    },
    {
      "epoch": 1.7396184062850728,
      "grad_norm": 0.7201360464096069,
      "learning_rate": 7.245292430477615e-05,
      "loss": 3.6137,
      "step": 3100
    },
    {
      "epoch": 1.7957351290684624,
      "grad_norm": 0.7597838640213013,
      "learning_rate": 7.120588602070084e-05,
      "loss": 3.5978,
      "step": 3200
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 0.7621691226959229,
      "learning_rate": 6.995884773662552e-05,
      "loss": 3.5693,
      "step": 3300
    },
    {
      "epoch": 1.9079685746352413,
      "grad_norm": 0.8080981373786926,
      "learning_rate": 6.871180945255021e-05,
      "loss": 3.562,
      "step": 3400
    },
    {
      "epoch": 1.964085297418631,
      "grad_norm": 0.7340370416641235,
      "learning_rate": 6.746477116847487e-05,
      "loss": 3.5435,
      "step": 3500
    },
    {
      "epoch": 1.964085297418631,
      "eval_accuracy": 0.38766734947221787,
      "eval_bleu": 0.13431823382301575,
      "eval_loss": 3.4940176010131836,
      "eval_perplexity": 32.917933518588185,
      "eval_runtime": 19.6051,
      "eval_samples_per_second": 58.199,
      "eval_steps_per_second": 0.918,
      "step": 3500
    },
    {
      "epoch": 2.0202020202020203,
      "grad_norm": 0.7243474721908569,
      "learning_rate": 6.621773288439955e-05,
      "loss": 3.5078,
      "step": 3600
    },
    {
      "epoch": 2.0763187429854097,
      "grad_norm": 0.8073205947875977,
      "learning_rate": 6.497069460032424e-05,
      "loss": 3.4528,
      "step": 3700
    },
    {
      "epoch": 2.132435465768799,
      "grad_norm": 0.7458230257034302,
      "learning_rate": 6.372365631624892e-05,
      "loss": 3.4305,
      "step": 3800
    },
    {
      "epoch": 2.1885521885521886,
      "grad_norm": 0.7389336228370667,
      "learning_rate": 6.247661803217359e-05,
      "loss": 3.4306,
      "step": 3900
    },
    {
      "epoch": 2.244668911335578,
      "grad_norm": 0.7879050374031067,
      "learning_rate": 6.122957974809826e-05,
      "loss": 3.4222,
      "step": 4000
    },
    {
      "epoch": 2.244668911335578,
      "eval_accuracy": 0.3935907090468737,
      "eval_bleu": 0.1343346102743112,
      "eval_loss": 3.4292232990264893,
      "eval_perplexity": 30.85267014217357,
      "eval_runtime": 19.5561,
      "eval_samples_per_second": 58.345,
      "eval_steps_per_second": 0.92,
      "step": 4000
    },
    {
      "epoch": 2.3007856341189674,
      "grad_norm": 0.7585355043411255,
      "learning_rate": 5.998254146402295e-05,
      "loss": 3.4101,
      "step": 4100
    },
    {
      "epoch": 2.356902356902357,
      "grad_norm": 0.7705636620521545,
      "learning_rate": 5.8735503179947625e-05,
      "loss": 3.3892,
      "step": 4200
    },
    {
      "epoch": 2.4130190796857462,
      "grad_norm": 0.7811786532402039,
      "learning_rate": 5.748846489587231e-05,
      "loss": 3.3858,
      "step": 4300
    },
    {
      "epoch": 2.4691358024691357,
      "grad_norm": 0.7358310222625732,
      "learning_rate": 5.624142661179699e-05,
      "loss": 3.3697,
      "step": 4400
    },
    {
      "epoch": 2.525252525252525,
      "grad_norm": 0.741968035697937,
      "learning_rate": 5.4994388327721666e-05,
      "loss": 3.3604,
      "step": 4500
    },
    {
      "epoch": 2.525252525252525,
      "eval_accuracy": 0.399036875783363,
      "eval_bleu": 0.14137470451886344,
      "eval_loss": 3.372802972793579,
      "eval_perplexity": 29.160147714819164,
      "eval_runtime": 12.9108,
      "eval_samples_per_second": 88.376,
      "eval_steps_per_second": 1.394,
      "step": 4500
    },
    {
      "epoch": 2.581369248035915,
      "grad_norm": 0.7442606091499329,
      "learning_rate": 5.374735004364634e-05,
      "loss": 3.3644,
      "step": 4600
    },
    {
      "epoch": 2.637485970819304,
      "grad_norm": 0.7207857966423035,
      "learning_rate": 5.250031175957102e-05,
      "loss": 3.3499,
      "step": 4700
    },
    {
      "epoch": 2.6936026936026938,
      "grad_norm": 0.7526234984397888,
      "learning_rate": 5.12532734754957e-05,
      "loss": 3.3445,
      "step": 4800
    },
    {
      "epoch": 2.749719416386083,
      "grad_norm": 0.7211634516716003,
      "learning_rate": 5.000623519142038e-05,
      "loss": 3.3345,
      "step": 4900
    },
    {
      "epoch": 2.8058361391694726,
      "grad_norm": 0.7400387525558472,
      "learning_rate": 4.8759196907345056e-05,
      "loss": 3.3288,
      "step": 5000
    },
    {
      "epoch": 2.8058361391694726,
      "eval_accuracy": 0.40383279231488217,
      "eval_bleu": 0.1380917557723357,
      "eval_loss": 3.3268959522247314,
      "eval_perplexity": 27.851754211207222,
      "eval_runtime": 19.4973,
      "eval_samples_per_second": 58.521,
      "eval_steps_per_second": 0.923,
      "step": 5000
    },
    {
      "epoch": 2.861952861952862,
      "grad_norm": 0.7265079021453857,
      "learning_rate": 4.751215862326974e-05,
      "loss": 3.3264,
      "step": 5100
    },
    {
      "epoch": 2.9180695847362514,
      "grad_norm": 0.7342014908790588,
      "learning_rate": 4.626512033919442e-05,
      "loss": 3.3048,
      "step": 5200
    },
    {
      "epoch": 2.974186307519641,
      "grad_norm": 0.7183738350868225,
      "learning_rate": 4.5018082055119096e-05,
      "loss": 3.3088,
      "step": 5300
    },
    {
      "epoch": 3.0303030303030303,
      "grad_norm": 0.7631810307502747,
      "learning_rate": 4.3771043771043774e-05,
      "loss": 3.256,
      "step": 5400
    },
    {
      "epoch": 3.0864197530864197,
      "grad_norm": 0.7537241578102112,
      "learning_rate": 4.252400548696845e-05,
      "loss": 3.2074,
      "step": 5500
    },
    {
      "epoch": 3.0864197530864197,
      "eval_accuracy": 0.4078876463598411,
      "eval_bleu": 0.14234868172058146,
      "eval_loss": 3.2887463569641113,
      "eval_perplexity": 26.80923337140352,
      "eval_runtime": 19.5139,
      "eval_samples_per_second": 58.471,
      "eval_steps_per_second": 0.922,
      "step": 5500
    },
    {
      "epoch": 3.142536475869809,
      "grad_norm": 0.7671452164649963,
      "learning_rate": 4.127696720289313e-05,
      "loss": 3.2138,
      "step": 5600
    },
    {
      "epoch": 3.1986531986531985,
      "grad_norm": 0.7754735350608826,
      "learning_rate": 4.002992891881781e-05,
      "loss": 3.2101,
      "step": 5700
    },
    {
      "epoch": 3.254769921436588,
      "grad_norm": 0.7768378257751465,
      "learning_rate": 3.8782890634742486e-05,
      "loss": 3.2069,
      "step": 5800
    },
    {
      "epoch": 3.3108866442199774,
      "grad_norm": 0.7553181052207947,
      "learning_rate": 3.7535852350667164e-05,
      "loss": 3.2078,
      "step": 5900
    },
    {
      "epoch": 3.3670033670033668,
      "grad_norm": 0.769962728023529,
      "learning_rate": 3.628881406659185e-05,
      "loss": 3.2007,
      "step": 6000
    },
    {
      "epoch": 3.3670033670033668,
      "eval_accuracy": 0.41151499730561675,
      "eval_bleu": 0.14635308836715238,
      "eval_loss": 3.2605247497558594,
      "eval_perplexity": 26.063210217931864,
      "eval_runtime": 19.4869,
      "eval_samples_per_second": 58.552,
      "eval_steps_per_second": 0.924,
      "step": 6000
    },
    {
      "epoch": 3.4231200897867566,
      "grad_norm": 0.7611728310585022,
      "learning_rate": 3.504177578251652e-05,
      "loss": 3.1805,
      "step": 6100
    },
    {
      "epoch": 3.479236812570146,
      "grad_norm": 0.7554565668106079,
      "learning_rate": 3.3794737498441205e-05,
      "loss": 3.1914,
      "step": 6200
    },
    {
      "epoch": 3.5353535353535355,
      "grad_norm": 0.7835715413093567,
      "learning_rate": 3.254769921436588e-05,
      "loss": 3.1813,
      "step": 6300
    },
    {
      "epoch": 3.591470258136925,
      "grad_norm": 0.7560538649559021,
      "learning_rate": 3.130066093029056e-05,
      "loss": 3.1827,
      "step": 6400
    },
    {
      "epoch": 3.6475869809203143,
      "grad_norm": 0.7631095051765442,
      "learning_rate": 3.0053622646215242e-05,
      "loss": 3.1787,
      "step": 6500
    },
    {
      "epoch": 3.6475869809203143,
      "eval_accuracy": 0.41400376785296633,
      "eval_bleu": 0.14275509438266887,
      "eval_loss": 3.23276686668396,
      "eval_perplexity": 25.349699265779424,
      "eval_runtime": 19.5684,
      "eval_samples_per_second": 58.308,
      "eval_steps_per_second": 0.92,
      "step": 6500
    },
    {
      "epoch": 3.7037037037037037,
      "grad_norm": 0.7613770365715027,
      "learning_rate": 2.880658436213992e-05,
      "loss": 3.171,
      "step": 6600
    },
    {
      "epoch": 3.759820426487093,
      "grad_norm": 0.7514590620994568,
      "learning_rate": 2.7559546078064598e-05,
      "loss": 3.1616,
      "step": 6700
    },
    {
      "epoch": 3.8159371492704826,
      "grad_norm": 0.7718783020973206,
      "learning_rate": 2.6312507793989276e-05,
      "loss": 3.1519,
      "step": 6800
    },
    {
      "epoch": 3.872053872053872,
      "grad_norm": 0.7550140023231506,
      "learning_rate": 2.5065469509913957e-05,
      "loss": 3.1439,
      "step": 6900
    },
    {
      "epoch": 3.9281705948372614,
      "grad_norm": 0.7449607253074646,
      "learning_rate": 2.3818431225838632e-05,
      "loss": 3.1529,
      "step": 7000
    },
    {
      "epoch": 3.9281705948372614,
      "eval_accuracy": 0.4165602192516897,
      "eval_bleu": 0.14252386333027117,
      "eval_loss": 3.208519220352173,
      "eval_perplexity": 24.742421011265083,
      "eval_runtime": 19.5498,
      "eval_samples_per_second": 58.364,
      "eval_steps_per_second": 0.921,
      "step": 7000
    },
    {
      "epoch": 3.984287317620651,
      "grad_norm": 0.7556090950965881,
      "learning_rate": 2.2571392941763313e-05,
      "loss": 3.1493,
      "step": 7100
    },
    {
      "epoch": 4.040404040404041,
      "grad_norm": 0.7740824222564697,
      "learning_rate": 2.132435465768799e-05,
      "loss": 3.0831,
      "step": 7200
    },
    {
      "epoch": 4.09652076318743,
      "grad_norm": 0.7735643982887268,
      "learning_rate": 2.007731637361267e-05,
      "loss": 3.0788,
      "step": 7300
    },
    {
      "epoch": 4.1526374859708195,
      "grad_norm": 0.8017794489860535,
      "learning_rate": 1.883027808953735e-05,
      "loss": 3.0808,
      "step": 7400
    },
    {
      "epoch": 4.2087542087542085,
      "grad_norm": 0.7887123227119446,
      "learning_rate": 1.758323980546203e-05,
      "loss": 3.0849,
      "step": 7500
    },
    {
      "epoch": 4.2087542087542085,
      "eval_accuracy": 0.4184484293330523,
      "eval_bleu": 0.14302496814022836,
      "eval_loss": 3.1920533180236816,
      "eval_perplexity": 24.33835054539309,
      "eval_runtime": 19.5741,
      "eval_samples_per_second": 58.291,
      "eval_steps_per_second": 0.92,
      "step": 7500
    },
    {
      "epoch": 4.264870931537598,
      "grad_norm": 0.7835370898246765,
      "learning_rate": 1.6336201521386706e-05,
      "loss": 3.0707,
      "step": 7600
    },
    {
      "epoch": 4.320987654320987,
      "grad_norm": 0.7777314782142639,
      "learning_rate": 1.5089163237311384e-05,
      "loss": 3.058,
      "step": 7700
    },
    {
      "epoch": 4.377104377104377,
      "grad_norm": 0.7813054919242859,
      "learning_rate": 1.3842124953236066e-05,
      "loss": 3.0678,
      "step": 7800
    },
    {
      "epoch": 4.433221099887767,
      "grad_norm": 0.7960292100906372,
      "learning_rate": 1.2595086669160744e-05,
      "loss": 3.0659,
      "step": 7900
    },
    {
      "epoch": 4.489337822671156,
      "grad_norm": 0.7847529053688049,
      "learning_rate": 1.1348048385085423e-05,
      "loss": 3.0471,
      "step": 8000
    },
    {
      "epoch": 4.489337822671156,
      "eval_accuracy": 0.42023297633826034,
      "eval_bleu": 0.14281001374264807,
      "eval_loss": 3.1795759201049805,
      "eval_perplexity": 24.036557969366196,
      "eval_runtime": 19.3711,
      "eval_samples_per_second": 58.902,
      "eval_steps_per_second": 0.929,
      "step": 8000
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 0.8106744885444641,
      "learning_rate": 1.0101010101010101e-05,
      "loss": 3.0626,
      "step": 8100
    },
    {
      "epoch": 4.601571268237935,
      "grad_norm": 0.7878694534301758,
      "learning_rate": 8.853971816934781e-06,
      "loss": 3.0555,
      "step": 8200
    },
    {
      "epoch": 4.657687991021325,
      "grad_norm": 0.7682947516441345,
      "learning_rate": 7.606933532859459e-06,
      "loss": 3.0621,
      "step": 8300
    },
    {
      "epoch": 4.713804713804714,
      "grad_norm": 0.7908104062080383,
      "learning_rate": 6.359895248784138e-06,
      "loss": 3.0546,
      "step": 8400
    },
    {
      "epoch": 4.7699214365881035,
      "grad_norm": 0.7882031798362732,
      "learning_rate": 5.112856964708817e-06,
      "loss": 3.0569,
      "step": 8500
    },
    {
      "epoch": 4.7699214365881035,
      "eval_accuracy": 0.42171338787210544,
      "eval_bleu": 0.14601329221778306,
      "eval_loss": 3.1670639514923096,
      "eval_perplexity": 23.737686941949445,
      "eval_runtime": 19.6936,
      "eval_samples_per_second": 57.938,
      "eval_steps_per_second": 0.914,
      "step": 8500
    },
    {
      "epoch": 4.8260381593714925,
      "grad_norm": 0.7835642099380493,
      "learning_rate": 3.865818680633495e-06,
      "loss": 3.0473,
      "step": 8600
    },
    {
      "epoch": 4.882154882154882,
      "grad_norm": 0.783760130405426,
      "learning_rate": 2.6187803965581742e-06,
      "loss": 3.0579,
      "step": 8700
    },
    {
      "epoch": 4.938271604938271,
      "grad_norm": 0.773915708065033,
      "learning_rate": 1.3717421124828533e-06,
      "loss": 3.0542,
      "step": 8800
    },
    {
      "epoch": 4.994388327721661,
      "grad_norm": 0.7747133374214172,
      "learning_rate": 1.2470382840753213e-07,
      "loss": 3.0443,
      "step": 8900
    },
    {
      "epoch": 5.0,
      "step": 8910,
      "total_flos": 1.0586630697202483e+18,
      "train_loss": 3.7611954097127005,
      "train_runtime": 27476.9125,
      "train_samples_per_second": 20.744,
      "train_steps_per_second": 0.324
    }
  ],
  "logging_steps": 100,
  "max_steps": 8910,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 2,
        "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": 1.0586630697202483e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}