{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.49996528018887576,
  "eval_steps": 1000,
  "global_step": 7200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 6.943962224845496e-05,
      "grad_norm": 9.8125,
      "learning_rate": 2e-06,
      "loss": 0.7188,
      "step": 1
    },
    {
      "epoch": 0.006943962224845497,
      "grad_norm": 0.1376953125,
      "learning_rate": 0.0002,
      "loss": 0.3907,
      "step": 100
    },
    {
      "epoch": 0.013887924449690994,
      "grad_norm": 0.0849609375,
      "learning_rate": 0.0004,
      "loss": 0.2341,
      "step": 200
    },
    {
      "epoch": 0.02083188667453649,
      "grad_norm": 0.146484375,
      "learning_rate": 0.0006,
      "loss": 0.2181,
      "step": 300
    },
    {
      "epoch": 0.027775848899381988,
      "grad_norm": 0.10546875,
      "learning_rate": 0.0008,
      "loss": 0.197,
      "step": 400
    },
    {
      "epoch": 0.034719811124227486,
      "grad_norm": 0.11376953125,
      "learning_rate": 0.001,
      "loss": 0.1695,
      "step": 500
    },
    {
      "epoch": 0.04166377334907298,
      "grad_norm": 0.0849609375,
      "learning_rate": 0.0012,
      "loss": 0.14,
      "step": 600
    },
    {
      "epoch": 0.04860773557391848,
      "grad_norm": 0.078125,
      "learning_rate": 0.0014,
      "loss": 0.1231,
      "step": 700
    },
    {
      "epoch": 0.055551697798763976,
      "grad_norm": 0.0712890625,
      "learning_rate": 0.0016,
      "loss": 0.1141,
      "step": 800
    },
    {
      "epoch": 0.06249566002360947,
      "grad_norm": 0.06298828125,
      "learning_rate": 0.0018000000000000002,
      "loss": 0.1077,
      "step": 900
    },
    {
      "epoch": 0.06943962224845497,
      "grad_norm": 0.0615234375,
      "learning_rate": 0.002,
      "loss": 0.1045,
      "step": 1000
    },
    {
      "epoch": 0.06943962224845497,
      "eval_covost2-en-de_loss": 1.4858413934707642,
      "eval_covost2-en-de_runtime": 32.1589,
      "eval_covost2-en-de_samples_per_second": 1.99,
      "eval_covost2-en-de_steps_per_second": 0.062,
      "step": 1000
    },
    {
      "epoch": 0.06943962224845497,
      "eval_covost2-zh-en_loss": 2.7152516841888428,
      "eval_covost2-zh-en_runtime": 31.3842,
      "eval_covost2-zh-en_samples_per_second": 2.039,
      "eval_covost2-zh-en_steps_per_second": 0.064,
      "step": 1000
    },
    {
      "epoch": 0.06943962224845497,
      "eval_peoplespeech-clean-transcription_loss": 2.0398874282836914,
      "eval_peoplespeech-clean-transcription_runtime": 32.088,
      "eval_peoplespeech-clean-transcription_samples_per_second": 1.995,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.062,
      "step": 1000
    },
    {
      "epoch": 0.07638358447330046,
      "grad_norm": 0.053955078125,
      "learning_rate": 0.001999725185109816,
      "loss": 0.101,
      "step": 1100
    },
    {
      "epoch": 0.08332754669814596,
      "grad_norm": 0.0517578125,
      "learning_rate": 0.0019989008914857113,
      "loss": 0.0956,
      "step": 1200
    },
    {
      "epoch": 0.09027150892299146,
      "grad_norm": 0.04443359375,
      "learning_rate": 0.00199752757218401,
      "loss": 0.0936,
      "step": 1300
    },
    {
      "epoch": 0.09721547114783696,
      "grad_norm": 0.0390625,
      "learning_rate": 0.001995605982021898,
      "loss": 0.0917,
      "step": 1400
    },
    {
      "epoch": 0.10415943337268245,
      "grad_norm": 0.0517578125,
      "learning_rate": 0.0019931371771625545,
      "loss": 0.0894,
      "step": 1500
    },
    {
      "epoch": 0.11110339559752795,
      "grad_norm": 0.0419921875,
      "learning_rate": 0.001990122514534651,
      "loss": 0.0868,
      "step": 1600
    },
    {
      "epoch": 0.11804735782237345,
      "grad_norm": 0.039306640625,
      "learning_rate": 0.0019865636510865464,
      "loss": 0.0861,
      "step": 1700
    },
    {
      "epoch": 0.12499132004721894,
      "grad_norm": 0.047119140625,
      "learning_rate": 0.001982462542875576,
      "loss": 0.0854,
      "step": 1800
    },
    {
      "epoch": 0.13193528227206444,
      "grad_norm": 0.0390625,
      "learning_rate": 0.001977821443992945,
      "loss": 0.0837,
      "step": 1900
    },
    {
      "epoch": 0.13887924449690994,
      "grad_norm": 0.04052734375,
      "learning_rate": 0.001972642905324813,
      "loss": 0.0818,
      "step": 2000
    },
    {
      "epoch": 0.13887924449690994,
      "eval_covost2-en-de_loss": 1.4137890338897705,
      "eval_covost2-en-de_runtime": 32.5714,
      "eval_covost2-en-de_samples_per_second": 1.965,
      "eval_covost2-en-de_steps_per_second": 0.061,
      "step": 2000
    },
    {
      "epoch": 0.13887924449690994,
      "eval_covost2-zh-en_loss": 2.667837381362915,
      "eval_covost2-zh-en_runtime": 31.1685,
      "eval_covost2-zh-en_samples_per_second": 2.053,
      "eval_covost2-zh-en_steps_per_second": 0.064,
      "step": 2000
    },
    {
      "epoch": 0.13887924449690994,
      "eval_peoplespeech-clean-transcription_loss": 1.835880160331726,
      "eval_peoplespeech-clean-transcription_runtime": 32.0265,
      "eval_peoplespeech-clean-transcription_samples_per_second": 1.998,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.062,
      "step": 2000
    },
    {
      "epoch": 0.14582320672175544,
      "grad_norm": 0.039794921875,
      "learning_rate": 0.0019669297731502505,
      "loss": 0.0813,
      "step": 2100
    },
    {
      "epoch": 0.15276716894660092,
      "grad_norm": 0.03515625,
      "learning_rate": 0.00196068518757684,
      "loss": 0.0811,
      "step": 2200
    },
    {
      "epoch": 0.15971113117144642,
      "grad_norm": 0.04443359375,
      "learning_rate": 0.001953912580814779,
      "loss": 0.0793,
      "step": 2300
    },
    {
      "epoch": 0.16665509339629192,
      "grad_norm": 0.037841796875,
      "learning_rate": 0.0019466156752904343,
      "loss": 0.0788,
      "step": 2400
    },
    {
      "epoch": 0.17359905562113742,
      "grad_norm": 0.04052734375,
      "learning_rate": 0.0019387984816003866,
      "loss": 0.0783,
      "step": 2500
    },
    {
      "epoch": 0.18054301784598292,
      "grad_norm": 0.03466796875,
      "learning_rate": 0.0019304652963070869,
      "loss": 0.0772,
      "step": 2600
    },
    {
      "epoch": 0.18748698007082842,
      "grad_norm": 0.036376953125,
      "learning_rate": 0.0019216206995773372,
      "loss": 0.0771,
      "step": 2700
    },
    {
      "epoch": 0.19443094229567393,
      "grad_norm": 0.0400390625,
      "learning_rate": 0.0019122695526648968,
      "loss": 0.0766,
      "step": 2800
    },
    {
      "epoch": 0.2013749045205194,
      "grad_norm": 0.0361328125,
      "learning_rate": 0.0019024169952385887,
      "loss": 0.0753,
      "step": 2900
    },
    {
      "epoch": 0.2083188667453649,
      "grad_norm": 0.03125,
      "learning_rate": 0.0018920684425573864,
      "loss": 0.075,
      "step": 3000
    },
    {
      "epoch": 0.2083188667453649,
      "eval_covost2-en-de_loss": 1.3828500509262085,
      "eval_covost2-en-de_runtime": 32.2972,
      "eval_covost2-en-de_samples_per_second": 1.982,
      "eval_covost2-en-de_steps_per_second": 0.062,
      "step": 3000
    },
    {
      "epoch": 0.2083188667453649,
      "eval_covost2-zh-en_loss": 2.646721839904785,
      "eval_covost2-zh-en_runtime": 31.0128,
      "eval_covost2-zh-en_samples_per_second": 2.064,
      "eval_covost2-zh-en_steps_per_second": 0.064,
      "step": 3000
    },
    {
      "epoch": 0.2083188667453649,
      "eval_peoplespeech-clean-transcription_loss": 1.773127555847168,
      "eval_peoplespeech-clean-transcription_runtime": 32.5192,
      "eval_peoplespeech-clean-transcription_samples_per_second": 1.968,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.062,
      "step": 3000
    },
    {
      "epoch": 0.2152628289702104,
      "grad_norm": 0.03369140625,
      "learning_rate": 0.0018812295824940284,
      "loss": 0.0743,
      "step": 3100
    },
    {
      "epoch": 0.2222067911950559,
      "grad_norm": 0.034912109375,
      "learning_rate": 0.0018699063724087904,
      "loss": 0.074,
      "step": 3200
    },
    {
      "epoch": 0.2291507534199014,
      "grad_norm": 0.033203125,
      "learning_rate": 0.0018581050358751443,
      "loss": 0.0742,
      "step": 3300
    },
    {
      "epoch": 0.2360947156447469,
      "grad_norm": 0.03857421875,
      "learning_rate": 0.0018458320592590974,
      "loss": 0.0742,
      "step": 3400
    },
    {
      "epoch": 0.24303867786959238,
      "grad_norm": 0.033935546875,
      "learning_rate": 0.0018330941881540914,
      "loss": 0.0728,
      "step": 3500
    },
    {
      "epoch": 0.24998264009443788,
      "grad_norm": 0.031005859375,
      "learning_rate": 0.0018198984236734246,
      "loss": 0.0728,
      "step": 3600
    },
    {
      "epoch": 0.2569266023192834,
      "grad_norm": 0.030029296875,
      "learning_rate": 0.0018062520186022297,
      "loss": 0.0714,
      "step": 3700
    },
    {
      "epoch": 0.2638705645441289,
      "grad_norm": 0.02734375,
      "learning_rate": 0.0017921624734111292,
      "loss": 0.071,
      "step": 3800
    },
    {
      "epoch": 0.2708145267689744,
      "grad_norm": 0.03271484375,
      "learning_rate": 0.001777637532133752,
      "loss": 0.0705,
      "step": 3900
    },
    {
      "epoch": 0.2777584889938199,
      "grad_norm": 0.035400390625,
      "learning_rate": 0.0017626851781103819,
      "loss": 0.0714,
      "step": 4000
    },
    {
      "epoch": 0.2777584889938199,
      "eval_covost2-en-de_loss": 1.3778624534606934,
      "eval_covost2-en-de_runtime": 32.6181,
      "eval_covost2-en-de_samples_per_second": 1.962,
      "eval_covost2-en-de_steps_per_second": 0.061,
      "step": 4000
    },
    {
      "epoch": 0.2777584889938199,
      "eval_covost2-zh-en_loss": 2.6438870429992676,
      "eval_covost2-zh-en_runtime": 31.4603,
      "eval_covost2-zh-en_samples_per_second": 2.034,
      "eval_covost2-zh-en_steps_per_second": 0.064,
      "step": 4000
    },
    {
      "epoch": 0.2777584889938199,
      "eval_peoplespeech-clean-transcription_loss": 1.7361584901809692,
      "eval_peoplespeech-clean-transcription_runtime": 32.442,
      "eval_peoplespeech-clean-transcription_samples_per_second": 1.973,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.062,
      "step": 4000
    },
    {
      "epoch": 0.2847024512186654,
      "grad_norm": 0.0281982421875,
      "learning_rate": 0.001747313629600077,
      "loss": 0.0713,
      "step": 4100
    },
    {
      "epoch": 0.2916464134435109,
      "grad_norm": 0.028076171875,
      "learning_rate": 0.001731531335263669,
      "loss": 0.0699,
      "step": 4200
    },
    {
      "epoch": 0.2985903756683564,
      "grad_norm": 0.0277099609375,
      "learning_rate": 0.0017153469695201276,
      "loss": 0.0702,
      "step": 4300
    },
    {
      "epoch": 0.30553433789320184,
      "grad_norm": 0.031982421875,
      "learning_rate": 0.0016987694277788418,
      "loss": 0.0692,
      "step": 4400
    },
    {
      "epoch": 0.31247830011804734,
      "grad_norm": 0.02880859375,
      "learning_rate": 0.001681807821550438,
      "loss": 0.0686,
      "step": 4500
    },
    {
      "epoch": 0.31942226234289284,
      "grad_norm": 0.0289306640625,
      "learning_rate": 0.0016644714734388218,
      "loss": 0.0698,
      "step": 4600
    },
    {
      "epoch": 0.32636622456773834,
      "grad_norm": 0.029541015625,
      "learning_rate": 0.0016467699120171987,
      "loss": 0.0683,
      "step": 4700
    },
    {
      "epoch": 0.33331018679258384,
      "grad_norm": 0.034423828125,
      "learning_rate": 0.001628712866590885,
      "loss": 0.0687,
      "step": 4800
    },
    {
      "epoch": 0.34025414901742934,
      "grad_norm": 0.0289306640625,
      "learning_rate": 0.0016103102618497923,
      "loss": 0.0684,
      "step": 4900
    },
    {
      "epoch": 0.34719811124227484,
      "grad_norm": 0.0263671875,
      "learning_rate": 0.0015915722124135226,
      "loss": 0.0681,
      "step": 5000
    },
    {
      "epoch": 0.34719811124227484,
      "eval_covost2-en-de_loss": 1.3711879253387451,
      "eval_covost2-en-de_runtime": 32.6293,
      "eval_covost2-en-de_samples_per_second": 1.961,
      "eval_covost2-en-de_steps_per_second": 0.061,
      "step": 5000
    },
    {
      "epoch": 0.34719811124227484,
      "eval_covost2-zh-en_loss": 2.6346511840820312,
      "eval_covost2-zh-en_runtime": 32.1513,
      "eval_covost2-zh-en_samples_per_second": 1.991,
      "eval_covost2-zh-en_steps_per_second": 0.062,
      "step": 5000
    },
    {
      "epoch": 0.34719811124227484,
      "eval_peoplespeech-clean-transcription_loss": 1.7350472211837769,
      "eval_peoplespeech-clean-transcription_runtime": 32.5813,
      "eval_peoplespeech-clean-transcription_samples_per_second": 1.964,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.061,
      "step": 5000
    },
    {
      "epoch": 0.35414207346712034,
      "grad_norm": 0.028076171875,
      "learning_rate": 0.001572509017272072,
      "loss": 0.0693,
      "step": 5100
    },
    {
      "epoch": 0.36108603569196585,
      "grad_norm": 0.030517578125,
      "learning_rate": 0.0015531311541251993,
      "loss": 0.0683,
      "step": 5200
    },
    {
      "epoch": 0.36802999791681135,
      "grad_norm": 0.031494140625,
      "learning_rate": 0.0015334492736235703,
      "loss": 0.0677,
      "step": 5300
    },
    {
      "epoch": 0.37497396014165685,
      "grad_norm": 0.0284423828125,
      "learning_rate": 0.0015134741935148419,
      "loss": 0.0669,
      "step": 5400
    },
    {
      "epoch": 0.38191792236650235,
      "grad_norm": 0.0302734375,
      "learning_rate": 0.0014932168926979072,
      "loss": 0.0669,
      "step": 5500
    },
    {
      "epoch": 0.38886188459134785,
      "grad_norm": 0.0260009765625,
      "learning_rate": 0.0014726885051885652,
      "loss": 0.0666,
      "step": 5600
    },
    {
      "epoch": 0.3958058468161933,
      "grad_norm": 0.033447265625,
      "learning_rate": 0.0014519003139999338,
      "loss": 0.0659,
      "step": 5700
    },
    {
      "epoch": 0.4027498090410388,
      "grad_norm": 0.0283203125,
      "learning_rate": 0.0014308637449409706,
      "loss": 0.0653,
      "step": 5800
    },
    {
      "epoch": 0.4096937712658843,
      "grad_norm": 0.0283203125,
      "learning_rate": 0.0014095903603365066,
      "loss": 0.0662,
      "step": 5900
    },
    {
      "epoch": 0.4166377334907298,
      "grad_norm": 0.0267333984375,
      "learning_rate": 0.0013880918526722496,
      "loss": 0.0665,
      "step": 6000
    },
    {
      "epoch": 0.4166377334907298,
      "eval_covost2-en-de_loss": 1.3651559352874756,
      "eval_covost2-en-de_runtime": 32.5621,
      "eval_covost2-en-de_samples_per_second": 1.965,
      "eval_covost2-en-de_steps_per_second": 0.061,
      "step": 6000
    },
    {
      "epoch": 0.4166377334907298,
      "eval_covost2-zh-en_loss": 2.6372551918029785,
      "eval_covost2-zh-en_runtime": 31.028,
      "eval_covost2-zh-en_samples_per_second": 2.063,
      "eval_covost2-zh-en_steps_per_second": 0.064,
      "step": 6000
    },
    {
      "epoch": 0.4166377334907298,
      "eval_peoplespeech-clean-transcription_loss": 1.7209596633911133,
      "eval_peoplespeech-clean-transcription_runtime": 32.6773,
      "eval_peoplespeech-clean-transcription_samples_per_second": 1.959,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.061,
      "step": 6000
    },
    {
      "epoch": 0.4235816957155753,
      "grad_norm": 0.028564453125,
      "learning_rate": 0.0013663800381682463,
      "loss": 0.0658,
      "step": 6100
    },
    {
      "epoch": 0.4305256579404208,
      "grad_norm": 0.0299072265625,
      "learning_rate": 0.0013444668502843329,
      "loss": 0.0657,
      "step": 6200
    },
    {
      "epoch": 0.4374696201652663,
      "grad_norm": 0.0296630859375,
      "learning_rate": 0.0013223643331611537,
      "loss": 0.0655,
      "step": 6300
    },
    {
      "epoch": 0.4444135823901118,
      "grad_norm": 0.0286865234375,
      "learning_rate": 0.001300084635000341,
      "loss": 0.0654,
      "step": 6400
    },
    {
      "epoch": 0.4513575446149573,
      "grad_norm": 0.028564453125,
      "learning_rate": 0.0012776400013875004,
      "loss": 0.0655,
      "step": 6500
    },
    {
      "epoch": 0.4583015068398028,
      "grad_norm": 0.030029296875,
      "learning_rate": 0.0012550427685616766,
      "loss": 0.0648,
      "step": 6600
    },
    {
      "epoch": 0.4652454690646483,
      "grad_norm": 0.037109375,
      "learning_rate": 0.0012323053566349834,
      "loss": 0.0654,
      "step": 6700
    },
    {
      "epoch": 0.4721894312894938,
      "grad_norm": 0.029296875,
      "learning_rate": 0.0012094402627661448,
      "loss": 0.0643,
      "step": 6800
    },
    {
      "epoch": 0.47913339351433926,
      "grad_norm": 0.030517578125,
      "learning_rate": 0.0011864600542916813,
      "loss": 0.0646,
      "step": 6900
    },
    {
      "epoch": 0.48607735573918476,
      "grad_norm": 0.037353515625,
      "learning_rate": 0.0011633773618185302,
      "loss": 0.0642,
      "step": 7000
    },
    {
      "epoch": 0.48607735573918476,
      "eval_covost2-en-de_loss": 1.3594859838485718,
      "eval_covost2-en-de_runtime": 32.6807,
      "eval_covost2-en-de_samples_per_second": 1.958,
      "eval_covost2-en-de_steps_per_second": 0.061,
      "step": 7000
    },
    {
      "epoch": 0.48607735573918476,
      "eval_covost2-zh-en_loss": 2.626713514328003,
      "eval_covost2-zh-en_runtime": 31.0228,
      "eval_covost2-zh-en_samples_per_second": 2.063,
      "eval_covost2-zh-en_steps_per_second": 0.064,
      "step": 7000
    },
    {
      "epoch": 0.48607735573918476,
      "eval_peoplespeech-clean-transcription_loss": 1.693739652633667,
      "eval_peoplespeech-clean-transcription_runtime": 31.9776,
      "eval_peoplespeech-clean-transcription_samples_per_second": 2.001,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.063,
      "step": 7000
    },
    {
      "epoch": 0.49302131796403026,
      "grad_norm": 0.02978515625,
      "learning_rate": 0.0011402048722818862,
      "loss": 0.0656,
      "step": 7100
    },
    {
      "epoch": 0.49996528018887576,
      "grad_norm": 0.0281982421875,
      "learning_rate": 0.0011169553219720827,
      "loss": 0.064,
      "step": 7200
    }
  ],
  "logging_steps": 100,
  "max_steps": 14400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3600,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8942929465072026e+18,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}