Md Mushfiqur Rahman
Upload with huggingface_hub
cd31e4f
raw
history blame
32.6 kB
{
"best_metric": 0.9102957175353575,
"best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/canine/canine-base-finetuned-masakhaner-conll_2003_en/checkpoint-15000",
"epoch": 34.16856492027335,
"global_step": 15000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.46,
"eval_accuracy_score": 0.905671048012793,
"eval_f1": 0.4031678791785615,
"eval_loss": 0.3176519572734833,
"eval_precision": 0.37412408135361475,
"eval_recall": 0.4371006389776358,
"eval_runtime": 25.4824,
"eval_samples_per_second": 127.539,
"eval_steps_per_second": 15.972,
"step": 200
},
{
"epoch": 0.91,
"eval_accuracy_score": 0.9367760053044191,
"eval_f1": 0.600400266844563,
"eval_loss": 0.20133168995380402,
"eval_precision": 0.5742935278030994,
"eval_recall": 0.6289936102236422,
"eval_runtime": 25.4664,
"eval_samples_per_second": 127.619,
"eval_steps_per_second": 15.982,
"step": 400
},
{
"epoch": 1.14,
"learning_rate": 4.865771812080537e-05,
"loss": 0.3604,
"step": 500
},
{
"epoch": 1.37,
"eval_accuracy_score": 0.9617379773002067,
"eval_f1": 0.766023166023166,
"eval_loss": 0.1343105286359787,
"eval_precision": 0.7414050822122571,
"eval_recall": 0.792332268370607,
"eval_runtime": 25.4735,
"eval_samples_per_second": 127.584,
"eval_steps_per_second": 15.977,
"step": 600
},
{
"epoch": 1.82,
"eval_accuracy_score": 0.9683294980303444,
"eval_f1": 0.7997698504027618,
"eval_loss": 0.11298365145921707,
"eval_precision": 0.7693726937269373,
"eval_recall": 0.832667731629393,
"eval_runtime": 25.4403,
"eval_samples_per_second": 127.75,
"eval_steps_per_second": 15.998,
"step": 800
},
{
"epoch": 2.28,
"learning_rate": 4.697986577181208e-05,
"loss": 0.1118,
"step": 1000
},
{
"epoch": 2.28,
"eval_accuracy_score": 0.9734389016732322,
"eval_f1": 0.8381742738589212,
"eval_loss": 0.09904822707176208,
"eval_precision": 0.8110177404295051,
"eval_recall": 0.8672124600638977,
"eval_runtime": 25.4063,
"eval_samples_per_second": 127.921,
"eval_steps_per_second": 16.02,
"step": 1000
},
{
"epoch": 2.73,
"eval_accuracy_score": 0.9739654432700183,
"eval_f1": 0.8414692624916082,
"eval_loss": 0.09652349352836609,
"eval_precision": 0.8095589592175678,
"eval_recall": 0.8759984025559105,
"eval_runtime": 25.4049,
"eval_samples_per_second": 127.928,
"eval_steps_per_second": 16.021,
"step": 1200
},
{
"epoch": 3.19,
"eval_accuracy_score": 0.9777682436912516,
"eval_f1": 0.8652216748768473,
"eval_loss": 0.09149660170078278,
"eval_precision": 0.853947880202256,
"eval_recall": 0.876797124600639,
"eval_runtime": 25.4531,
"eval_samples_per_second": 127.686,
"eval_steps_per_second": 15.99,
"step": 1400
},
{
"epoch": 3.42,
"learning_rate": 4.530201342281879e-05,
"loss": 0.0611,
"step": 1500
},
{
"epoch": 3.64,
"eval_accuracy_score": 0.9783922929911463,
"eval_f1": 0.8685501275259958,
"eval_loss": 0.08810298144817352,
"eval_precision": 0.8536444273042808,
"eval_recall": 0.8839856230031949,
"eval_runtime": 25.471,
"eval_samples_per_second": 127.596,
"eval_steps_per_second": 15.979,
"step": 1600
},
{
"epoch": 4.1,
"eval_accuracy_score": 0.9792113576972581,
"eval_f1": 0.8704247406537483,
"eval_loss": 0.08606501668691635,
"eval_precision": 0.8535508637236084,
"eval_recall": 0.887979233226837,
"eval_runtime": 25.4698,
"eval_samples_per_second": 127.602,
"eval_steps_per_second": 15.98,
"step": 1800
},
{
"epoch": 4.56,
"learning_rate": 4.36241610738255e-05,
"loss": 0.0382,
"step": 2000
},
{
"epoch": 4.56,
"eval_accuracy_score": 0.9797574008346659,
"eval_f1": 0.875222816399287,
"eval_loss": 0.08422768861055374,
"eval_precision": 0.8681728880157171,
"eval_recall": 0.882388178913738,
"eval_runtime": 25.4807,
"eval_samples_per_second": 127.547,
"eval_steps_per_second": 15.973,
"step": 2000
},
{
"epoch": 5.01,
"eval_accuracy_score": 0.9801084285658567,
"eval_f1": 0.8783047826517477,
"eval_loss": 0.08461492508649826,
"eval_precision": 0.8711451581221764,
"eval_recall": 0.8855830670926518,
"eval_runtime": 25.4915,
"eval_samples_per_second": 127.494,
"eval_steps_per_second": 15.966,
"step": 2200
},
{
"epoch": 5.47,
"eval_accuracy_score": 0.9802839424314521,
"eval_f1": 0.8772208436724566,
"eval_loss": 0.0915483608841896,
"eval_precision": 0.872113676731794,
"eval_recall": 0.882388178913738,
"eval_runtime": 25.5288,
"eval_samples_per_second": 127.307,
"eval_steps_per_second": 15.943,
"step": 2400
},
{
"epoch": 5.69,
"learning_rate": 4.194630872483222e-05,
"loss": 0.0269,
"step": 2500
},
{
"epoch": 5.92,
"eval_accuracy_score": 0.9790358438316628,
"eval_f1": 0.8738092899931258,
"eval_loss": 0.08523886650800705,
"eval_precision": 0.8597101449275363,
"eval_recall": 0.8883785942492013,
"eval_runtime": 25.4488,
"eval_samples_per_second": 127.707,
"eval_steps_per_second": 15.993,
"step": 2600
},
{
"epoch": 6.38,
"eval_accuracy_score": 0.980556964000156,
"eval_f1": 0.8805166846071044,
"eval_loss": 0.0843101516366005,
"eval_precision": 0.8633659566302053,
"eval_recall": 0.8983626198083067,
"eval_runtime": 25.4351,
"eval_samples_per_second": 127.776,
"eval_steps_per_second": 16.002,
"step": 2800
},
{
"epoch": 6.83,
"learning_rate": 4.026845637583892e-05,
"loss": 0.0195,
"step": 3000
},
{
"epoch": 6.83,
"eval_accuracy_score": 0.9802449393502086,
"eval_f1": 0.8802047445614726,
"eval_loss": 0.09137295186519623,
"eval_precision": 0.8679867986798679,
"eval_recall": 0.8927715654952076,
"eval_runtime": 25.4199,
"eval_samples_per_second": 127.853,
"eval_steps_per_second": 16.011,
"step": 3000
},
{
"epoch": 7.29,
"eval_accuracy_score": 0.9824096103592184,
"eval_f1": 0.8886048569462195,
"eval_loss": 0.08367849141359329,
"eval_precision": 0.8752663180321518,
"eval_recall": 0.9023562300319489,
"eval_runtime": 25.6211,
"eval_samples_per_second": 126.849,
"eval_steps_per_second": 15.885,
"step": 3200
},
{
"epoch": 7.74,
"eval_accuracy_score": 0.9814540348687546,
"eval_f1": 0.886024025783768,
"eval_loss": 0.08719700574874878,
"eval_precision": 0.8671382144905372,
"eval_recall": 0.9057507987220448,
"eval_runtime": 25.3936,
"eval_samples_per_second": 127.985,
"eval_steps_per_second": 16.028,
"step": 3400
},
{
"epoch": 7.97,
"learning_rate": 3.859060402684564e-05,
"loss": 0.0136,
"step": 3500
},
{
"epoch": 8.2,
"eval_accuracy_score": 0.9823316041967315,
"eval_f1": 0.8920094936708859,
"eval_loss": 0.0911451056599617,
"eval_precision": 0.8836206896551724,
"eval_recall": 0.9005591054313099,
"eval_runtime": 25.4627,
"eval_samples_per_second": 127.638,
"eval_steps_per_second": 15.984,
"step": 3600
},
{
"epoch": 8.66,
"eval_accuracy_score": 0.9830921642809782,
"eval_f1": 0.8962131837307153,
"eval_loss": 0.08473534882068634,
"eval_precision": 0.8992762364294331,
"eval_recall": 0.8931709265175719,
"eval_runtime": 25.4742,
"eval_samples_per_second": 127.58,
"eval_steps_per_second": 15.977,
"step": 3800
},
{
"epoch": 9.11,
"learning_rate": 3.6912751677852356e-05,
"loss": 0.011,
"step": 4000
},
{
"epoch": 9.11,
"eval_accuracy_score": 0.9824096103592184,
"eval_f1": 0.8943864878291108,
"eval_loss": 0.08603190630674362,
"eval_precision": 0.8900533913387384,
"eval_recall": 0.8987619808306709,
"eval_runtime": 25.451,
"eval_samples_per_second": 127.696,
"eval_steps_per_second": 15.991,
"step": 4000
},
{
"epoch": 9.57,
"eval_accuracy_score": 0.9822535980342447,
"eval_f1": 0.8913835548990644,
"eval_loss": 0.09200570732355118,
"eval_precision": 0.8793471925393433,
"eval_recall": 0.9037539936102237,
"eval_runtime": 25.5623,
"eval_samples_per_second": 127.141,
"eval_steps_per_second": 15.922,
"step": 4200
},
{
"epoch": 10.02,
"eval_accuracy_score": 0.9827801396310308,
"eval_f1": 0.8969183814118342,
"eval_loss": 0.08794757723808289,
"eval_precision": 0.8846377937463585,
"eval_recall": 0.9095447284345048,
"eval_runtime": 25.5902,
"eval_samples_per_second": 127.002,
"eval_steps_per_second": 15.905,
"step": 4400
},
{
"epoch": 10.25,
"learning_rate": 3.523489932885906e-05,
"loss": 0.0085,
"step": 4500
},
{
"epoch": 10.48,
"eval_accuracy_score": 0.9820000780061625,
"eval_f1": 0.890210067380103,
"eval_loss": 0.09557678550481796,
"eval_precision": 0.8835562549173879,
"eval_recall": 0.896964856230032,
"eval_runtime": 25.5311,
"eval_samples_per_second": 127.296,
"eval_steps_per_second": 15.941,
"step": 4600
},
{
"epoch": 10.93,
"eval_accuracy_score": 0.9825266196029486,
"eval_f1": 0.8952569169960475,
"eval_loss": 0.0915537029504776,
"eval_precision": 0.886150234741784,
"eval_recall": 0.9045527156549521,
"eval_runtime": 25.5326,
"eval_samples_per_second": 127.288,
"eval_steps_per_second": 15.94,
"step": 4800
},
{
"epoch": 11.39,
"learning_rate": 3.3557046979865775e-05,
"loss": 0.0077,
"step": 5000
},
{
"epoch": 11.39,
"eval_accuracy_score": 0.9818635672218106,
"eval_f1": 0.8888670917116234,
"eval_loss": 0.09687037765979767,
"eval_precision": 0.8735299787931367,
"eval_recall": 0.9047523961661342,
"eval_runtime": 25.5023,
"eval_samples_per_second": 127.44,
"eval_steps_per_second": 15.959,
"step": 5000
},
{
"epoch": 11.85,
"eval_accuracy_score": 0.9833846873903038,
"eval_f1": 0.9005432098765432,
"eval_loss": 0.08850205689668655,
"eval_precision": 0.8909517295290209,
"eval_recall": 0.9103434504792333,
"eval_runtime": 25.8091,
"eval_samples_per_second": 125.925,
"eval_steps_per_second": 15.77,
"step": 5200
},
{
"epoch": 12.3,
"eval_accuracy_score": 0.9832676781465736,
"eval_f1": 0.8994011976047904,
"eval_loss": 0.09829097986221313,
"eval_precision": 0.8990422984836393,
"eval_recall": 0.8997603833865815,
"eval_runtime": 25.5033,
"eval_samples_per_second": 127.434,
"eval_steps_per_second": 15.959,
"step": 5400
},
{
"epoch": 12.53,
"learning_rate": 3.1879194630872485e-05,
"loss": 0.0064,
"step": 5500
},
{
"epoch": 12.76,
"eval_accuracy_score": 0.9810054994344554,
"eval_f1": 0.8866424575522622,
"eval_loss": 0.1039794534444809,
"eval_precision": 0.8718394132406871,
"eval_recall": 0.9019568690095847,
"eval_runtime": 25.823,
"eval_samples_per_second": 125.857,
"eval_steps_per_second": 15.761,
"step": 5600
},
{
"epoch": 13.21,
"eval_accuracy_score": 0.9827606380904091,
"eval_f1": 0.8954599761051374,
"eval_loss": 0.09817508608102798,
"eval_precision": 0.8929706115965051,
"eval_recall": 0.8979632587859425,
"eval_runtime": 25.5457,
"eval_samples_per_second": 127.223,
"eval_steps_per_second": 15.932,
"step": 5800
},
{
"epoch": 13.67,
"learning_rate": 3.02013422818792e-05,
"loss": 0.0053,
"step": 6000
},
{
"epoch": 13.67,
"eval_accuracy_score": 0.9824291118998401,
"eval_f1": 0.8904041901373655,
"eval_loss": 0.10226656496524811,
"eval_precision": 0.8814322050479358,
"eval_recall": 0.8995607028753994,
"eval_runtime": 25.5455,
"eval_samples_per_second": 127.224,
"eval_steps_per_second": 15.932,
"step": 6000
},
{
"epoch": 14.12,
"eval_accuracy_score": 0.9825461211435703,
"eval_f1": 0.8980195093112622,
"eval_loss": 0.10106658190488815,
"eval_precision": 0.8864034234584711,
"eval_recall": 0.909944089456869,
"eval_runtime": 25.4581,
"eval_samples_per_second": 127.661,
"eval_steps_per_second": 15.987,
"step": 6200
},
{
"epoch": 14.58,
"eval_accuracy_score": 0.9829361519560045,
"eval_f1": 0.8955665024630541,
"eval_loss": 0.10531075298786163,
"eval_precision": 0.8838973162193698,
"eval_recall": 0.9075479233226837,
"eval_runtime": 25.4921,
"eval_samples_per_second": 127.491,
"eval_steps_per_second": 15.966,
"step": 6400
},
{
"epoch": 14.81,
"learning_rate": 2.8523489932885905e-05,
"loss": 0.0047,
"step": 6500
},
{
"epoch": 15.03,
"eval_accuracy_score": 0.9832091735247085,
"eval_f1": 0.9004636480220973,
"eval_loss": 0.0937289446592331,
"eval_precision": 0.8898420744784559,
"eval_recall": 0.9113418530351438,
"eval_runtime": 25.4989,
"eval_samples_per_second": 127.456,
"eval_steps_per_second": 15.961,
"step": 6600
},
{
"epoch": 15.49,
"eval_accuracy_score": 0.9830141581184914,
"eval_f1": 0.8983334976826743,
"eval_loss": 0.10006673634052277,
"eval_precision": 0.8873952854081434,
"eval_recall": 0.9095447284345048,
"eval_runtime": 25.5372,
"eval_samples_per_second": 127.265,
"eval_steps_per_second": 15.938,
"step": 6800
},
{
"epoch": 15.95,
"learning_rate": 2.6845637583892618e-05,
"loss": 0.0037,
"step": 7000
},
{
"epoch": 15.95,
"eval_accuracy_score": 0.9821170872498928,
"eval_f1": 0.8939078751857356,
"eval_loss": 0.10096853971481323,
"eval_precision": 0.886966778061726,
"eval_recall": 0.9009584664536742,
"eval_runtime": 25.5089,
"eval_samples_per_second": 127.406,
"eval_steps_per_second": 15.955,
"step": 7000
},
{
"epoch": 16.4,
"eval_accuracy_score": 0.9839112289870899,
"eval_f1": 0.9025406907502977,
"eval_loss": 0.09699645638465881,
"eval_precision": 0.8971981057616417,
"eval_recall": 0.9079472843450479,
"eval_runtime": 25.3546,
"eval_samples_per_second": 128.182,
"eval_steps_per_second": 16.052,
"step": 7200
},
{
"epoch": 16.86,
"eval_accuracy_score": 0.9833846873903038,
"eval_f1": 0.9011811023622048,
"eval_loss": 0.10188570618629456,
"eval_precision": 0.8885869565217391,
"eval_recall": 0.9141373801916933,
"eval_runtime": 25.3306,
"eval_samples_per_second": 128.303,
"eval_steps_per_second": 16.068,
"step": 7400
},
{
"epoch": 17.08,
"learning_rate": 2.516778523489933e-05,
"loss": 0.003,
"step": 7500
},
{
"epoch": 17.31,
"eval_accuracy_score": 0.9829556534966263,
"eval_f1": 0.9035664474995019,
"eval_loss": 0.11874072253704071,
"eval_precision": 0.9015904572564613,
"eval_recall": 0.9055511182108626,
"eval_runtime": 25.391,
"eval_samples_per_second": 127.998,
"eval_steps_per_second": 16.029,
"step": 7600
},
{
"epoch": 17.77,
"eval_accuracy_score": 0.9826241273060572,
"eval_f1": 0.8982106189498388,
"eval_loss": 0.10930322110652924,
"eval_precision": 0.880053650124545,
"eval_recall": 0.917132587859425,
"eval_runtime": 25.3359,
"eval_samples_per_second": 128.277,
"eval_steps_per_second": 16.064,
"step": 7800
},
{
"epoch": 18.22,
"learning_rate": 2.348993288590604e-05,
"loss": 0.0023,
"step": 8000
},
{
"epoch": 18.22,
"eval_accuracy_score": 0.9830141581184914,
"eval_f1": 0.8999901273570936,
"eval_loss": 0.10627970844507217,
"eval_precision": 0.8900605350517478,
"eval_recall": 0.9101437699680511,
"eval_runtime": 25.3382,
"eval_samples_per_second": 128.265,
"eval_steps_per_second": 16.063,
"step": 8000
},
{
"epoch": 18.68,
"eval_accuracy_score": 0.983774718202738,
"eval_f1": 0.9028480698620622,
"eval_loss": 0.09893312305212021,
"eval_precision": 0.8974156638390215,
"eval_recall": 0.9083466453674122,
"eval_runtime": 25.3736,
"eval_samples_per_second": 128.086,
"eval_steps_per_second": 16.04,
"step": 8200
},
{
"epoch": 19.13,
"eval_accuracy_score": 0.9830921642809782,
"eval_f1": 0.9000591366055589,
"eval_loss": 0.11214980483055115,
"eval_precision": 0.8886726352666408,
"eval_recall": 0.911741214057508,
"eval_runtime": 25.4824,
"eval_samples_per_second": 127.539,
"eval_steps_per_second": 15.972,
"step": 8400
},
{
"epoch": 19.36,
"learning_rate": 2.181208053691275e-05,
"loss": 0.0022,
"step": 8500
},
{
"epoch": 19.59,
"eval_accuracy_score": 0.9835406997152775,
"eval_f1": 0.8996444093243776,
"eval_loss": 0.10627951472997665,
"eval_precision": 0.8901485535574668,
"eval_recall": 0.9093450479233227,
"eval_runtime": 25.3697,
"eval_samples_per_second": 128.105,
"eval_steps_per_second": 16.043,
"step": 8600
},
{
"epoch": 20.05,
"eval_accuracy_score": 0.9832676781465736,
"eval_f1": 0.9012602957229334,
"eval_loss": 0.1101519763469696,
"eval_precision": 0.8958374432826988,
"eval_recall": 0.9067492012779552,
"eval_runtime": 25.36,
"eval_samples_per_second": 128.154,
"eval_steps_per_second": 16.049,
"step": 8800
},
{
"epoch": 20.5,
"learning_rate": 2.013422818791946e-05,
"loss": 0.0023,
"step": 9000
},
{
"epoch": 20.5,
"eval_accuracy_score": 0.9834626935527907,
"eval_f1": 0.8975216365066876,
"eval_loss": 0.10789068043231964,
"eval_precision": 0.8843023255813953,
"eval_recall": 0.9111421725239617,
"eval_runtime": 25.3261,
"eval_samples_per_second": 128.326,
"eval_steps_per_second": 16.07,
"step": 9000
},
{
"epoch": 20.96,
"eval_accuracy_score": 0.9833456843090604,
"eval_f1": 0.900358708648864,
"eval_loss": 0.11746969074010849,
"eval_precision": 0.8985680190930787,
"eval_recall": 0.9021565495207667,
"eval_runtime": 25.4056,
"eval_samples_per_second": 127.924,
"eval_steps_per_second": 16.02,
"step": 9200
},
{
"epoch": 21.41,
"eval_accuracy_score": 0.9833651858496821,
"eval_f1": 0.8977183320220299,
"eval_loss": 0.1041969433426857,
"eval_precision": 0.8844961240310077,
"eval_recall": 0.9113418530351438,
"eval_runtime": 25.4178,
"eval_samples_per_second": 127.863,
"eval_steps_per_second": 16.012,
"step": 9400
},
{
"epoch": 21.64,
"learning_rate": 1.8456375838926178e-05,
"loss": 0.0022,
"step": 9500
},
{
"epoch": 21.87,
"eval_accuracy_score": 0.9844572721244979,
"eval_f1": 0.905930267209695,
"eval_loss": 0.10449470579624176,
"eval_precision": 0.901363905910259,
"eval_recall": 0.9105431309904153,
"eval_runtime": 25.3884,
"eval_samples_per_second": 128.011,
"eval_steps_per_second": 16.031,
"step": 9600
},
{
"epoch": 22.32,
"eval_accuracy_score": 0.9839502320683334,
"eval_f1": 0.9034056960764787,
"eval_loss": 0.10690420866012573,
"eval_precision": 0.901072705601907,
"eval_recall": 0.9057507987220448,
"eval_runtime": 25.4089,
"eval_samples_per_second": 127.908,
"eval_steps_per_second": 16.018,
"step": 9800
},
{
"epoch": 22.78,
"learning_rate": 1.6778523489932888e-05,
"loss": 0.0019,
"step": 10000
},
{
"epoch": 22.78,
"eval_accuracy_score": 0.9835992043371427,
"eval_f1": 0.9027144838517931,
"eval_loss": 0.10763387382030487,
"eval_precision": 0.8957923712151002,
"eval_recall": 0.9097444089456869,
"eval_runtime": 25.3298,
"eval_samples_per_second": 128.307,
"eval_steps_per_second": 16.068,
"step": 10000
},
{
"epoch": 23.23,
"eval_accuracy_score": 0.9846522875307149,
"eval_f1": 0.9070299294024062,
"eval_loss": 0.10220099985599518,
"eval_precision": 0.9033471974648445,
"eval_recall": 0.9107428115015974,
"eval_runtime": 25.3799,
"eval_samples_per_second": 128.054,
"eval_steps_per_second": 16.036,
"step": 10200
},
{
"epoch": 23.69,
"eval_accuracy_score": 0.9840282382308202,
"eval_f1": 0.9080059671805072,
"eval_loss": 0.11143822968006134,
"eval_precision": 0.9044977214186646,
"eval_recall": 0.9115415335463258,
"eval_runtime": 25.375,
"eval_samples_per_second": 128.079,
"eval_steps_per_second": 16.039,
"step": 10400
},
{
"epoch": 23.92,
"learning_rate": 1.51006711409396e-05,
"loss": 0.0013,
"step": 10500
},
{
"epoch": 24.15,
"eval_accuracy_score": 0.9839502320683334,
"eval_f1": 0.9035462401907223,
"eval_loss": 0.11336545646190643,
"eval_precision": 0.8989918956315477,
"eval_recall": 0.90814696485623,
"eval_runtime": 25.4002,
"eval_samples_per_second": 127.952,
"eval_steps_per_second": 16.024,
"step": 10600
},
{
"epoch": 24.6,
"eval_accuracy_score": 0.9843012597995242,
"eval_f1": 0.9057239057239057,
"eval_loss": 0.10835720598697662,
"eval_precision": 0.8984282907662082,
"eval_recall": 0.9131389776357828,
"eval_runtime": 25.3998,
"eval_samples_per_second": 127.954,
"eval_steps_per_second": 16.024,
"step": 10800
},
{
"epoch": 25.06,
"learning_rate": 1.3422818791946309e-05,
"loss": 0.0009,
"step": 11000
},
{
"epoch": 25.06,
"eval_accuracy_score": 0.984379265962011,
"eval_f1": 0.9084597837945055,
"eval_loss": 0.10998024046421051,
"eval_precision": 0.9024630541871921,
"eval_recall": 0.9145367412140575,
"eval_runtime": 25.3705,
"eval_samples_per_second": 128.102,
"eval_steps_per_second": 16.042,
"step": 11000
},
{
"epoch": 25.51,
"eval_accuracy_score": 0.9838137212839815,
"eval_f1": 0.9037773359840955,
"eval_loss": 0.11049986630678177,
"eval_precision": 0.8998416468725258,
"eval_recall": 0.9077476038338658,
"eval_runtime": 25.3707,
"eval_samples_per_second": 128.1,
"eval_steps_per_second": 16.042,
"step": 11200
},
{
"epoch": 25.97,
"eval_accuracy_score": 0.9844767736651195,
"eval_f1": 0.9064790964929661,
"eval_loss": 0.10651768743991852,
"eval_precision": 0.8995281163979552,
"eval_recall": 0.9135383386581469,
"eval_runtime": 25.4896,
"eval_samples_per_second": 127.503,
"eval_steps_per_second": 15.967,
"step": 11400
},
{
"epoch": 26.2,
"learning_rate": 1.174496644295302e-05,
"loss": 0.0012,
"step": 11500
},
{
"epoch": 26.42,
"eval_accuracy_score": 0.9842817582589024,
"eval_f1": 0.9092719848620655,
"eval_loss": 0.1079045906662941,
"eval_precision": 0.9070137095171865,
"eval_recall": 0.9115415335463258,
"eval_runtime": 25.3954,
"eval_samples_per_second": 127.976,
"eval_steps_per_second": 16.027,
"step": 11600
},
{
"epoch": 26.88,
"eval_accuracy_score": 0.984184250555794,
"eval_f1": 0.9041501976284585,
"eval_loss": 0.10910345613956451,
"eval_precision": 0.8949530516431925,
"eval_recall": 0.9135383386581469,
"eval_runtime": 25.3926,
"eval_samples_per_second": 127.99,
"eval_steps_per_second": 16.028,
"step": 11800
},
{
"epoch": 27.33,
"learning_rate": 1.006711409395973e-05,
"loss": 0.0007,
"step": 12000
},
{
"epoch": 27.33,
"eval_accuracy_score": 0.9837942197433597,
"eval_f1": 0.9048899512000796,
"eval_loss": 0.11004339158535004,
"eval_precision": 0.9026425591098748,
"eval_recall": 0.9071485623003195,
"eval_runtime": 25.4253,
"eval_samples_per_second": 127.826,
"eval_steps_per_second": 16.008,
"step": 12000
},
{
"epoch": 27.79,
"eval_accuracy_score": 0.9841452474745505,
"eval_f1": 0.9081166898194086,
"eval_loss": 0.10656478255987167,
"eval_precision": 0.9025641025641026,
"eval_recall": 0.9137380191693291,
"eval_runtime": 25.3771,
"eval_samples_per_second": 128.068,
"eval_steps_per_second": 16.038,
"step": 12200
},
{
"epoch": 28.25,
"eval_accuracy_score": 0.9843597644213893,
"eval_f1": 0.9104314038059181,
"eval_loss": 0.10543405264616013,
"eval_precision": 0.9085305229667926,
"eval_recall": 0.9123402555910544,
"eval_runtime": 25.302,
"eval_samples_per_second": 128.448,
"eval_steps_per_second": 16.086,
"step": 12400
},
{
"epoch": 28.47,
"learning_rate": 8.389261744966444e-06,
"loss": 0.0008,
"step": 12500
},
{
"epoch": 28.7,
"eval_accuracy_score": 0.9839892351495768,
"eval_f1": 0.9044737625235593,
"eval_loss": 0.10808327049016953,
"eval_precision": 0.8986792824758526,
"eval_recall": 0.9103434504792333,
"eval_runtime": 25.3718,
"eval_samples_per_second": 128.095,
"eval_steps_per_second": 16.041,
"step": 12600
},
{
"epoch": 29.16,
"eval_accuracy_score": 0.9845937829088498,
"eval_f1": 0.9098848749503771,
"eval_loss": 0.10423844307661057,
"eval_precision": 0.904498816101026,
"eval_recall": 0.9153354632587859,
"eval_runtime": 25.3457,
"eval_samples_per_second": 128.227,
"eval_steps_per_second": 16.058,
"step": 12800
},
{
"epoch": 29.61,
"learning_rate": 6.7114093959731546e-06,
"loss": 0.0007,
"step": 13000
},
{
"epoch": 29.61,
"eval_accuracy_score": 0.9845742813682281,
"eval_f1": 0.9097922848664688,
"eval_loss": 0.10766186565160751,
"eval_precision": 0.9014112112896904,
"eval_recall": 0.9183306709265175,
"eval_runtime": 25.3084,
"eval_samples_per_second": 128.416,
"eval_steps_per_second": 16.082,
"step": 13000
},
{
"epoch": 30.07,
"eval_accuracy_score": 0.9844962752057412,
"eval_f1": 0.9077182874739247,
"eval_loss": 0.10705368965864182,
"eval_precision": 0.9031429136192923,
"eval_recall": 0.9123402555910544,
"eval_runtime": 25.4717,
"eval_samples_per_second": 127.593,
"eval_steps_per_second": 15.979,
"step": 13200
},
{
"epoch": 30.52,
"eval_accuracy_score": 0.9840087366901985,
"eval_f1": 0.9052005943536403,
"eval_loss": 0.10472096502780914,
"eval_precision": 0.8981718104973462,
"eval_recall": 0.9123402555910544,
"eval_runtime": 25.4157,
"eval_samples_per_second": 127.874,
"eval_steps_per_second": 16.014,
"step": 13400
},
{
"epoch": 30.75,
"learning_rate": 5.033557046979865e-06,
"loss": 0.0006,
"step": 13500
},
{
"epoch": 30.98,
"eval_accuracy_score": 0.9845937829088498,
"eval_f1": 0.9085492999702115,
"eval_loss": 0.10349733382463455,
"eval_precision": 0.9036144578313253,
"eval_recall": 0.9135383386581469,
"eval_runtime": 25.5031,
"eval_samples_per_second": 127.435,
"eval_steps_per_second": 15.959,
"step": 13600
},
{
"epoch": 31.44,
"eval_accuracy_score": 0.9849058075587972,
"eval_f1": 0.908856746582128,
"eval_loss": 0.10392281413078308,
"eval_precision": 0.9018875344081794,
"eval_recall": 0.9159345047923323,
"eval_runtime": 25.3529,
"eval_samples_per_second": 128.191,
"eval_steps_per_second": 16.053,
"step": 13800
},
{
"epoch": 31.89,
"learning_rate": 3.3557046979865773e-06,
"loss": 0.0005,
"step": 14000
},
{
"epoch": 31.89,
"eval_accuracy_score": 0.9849253090994189,
"eval_f1": 0.909863609408974,
"eval_loss": 0.10778042674064636,
"eval_precision": 0.9007827788649706,
"eval_recall": 0.919129392971246,
"eval_runtime": 25.3322,
"eval_samples_per_second": 128.295,
"eval_steps_per_second": 16.066,
"step": 14000
},
{
"epoch": 32.35,
"eval_accuracy_score": 0.9845742813682281,
"eval_f1": 0.9082278481012659,
"eval_loss": 0.10826905816793442,
"eval_precision": 0.8996865203761756,
"eval_recall": 0.9169329073482428,
"eval_runtime": 25.3345,
"eval_samples_per_second": 128.283,
"eval_steps_per_second": 16.065,
"step": 14200
},
{
"epoch": 32.8,
"eval_accuracy_score": 0.9846522875307149,
"eval_f1": 0.9092885375494072,
"eval_loss": 0.10940501093864441,
"eval_precision": 0.900039123630673,
"eval_recall": 0.9187300319488818,
"eval_runtime": 25.4019,
"eval_samples_per_second": 127.943,
"eval_steps_per_second": 16.022,
"step": 14400
},
{
"epoch": 33.03,
"learning_rate": 1.6778523489932886e-06,
"loss": 0.0003,
"step": 14500
},
{
"epoch": 33.26,
"eval_accuracy_score": 0.9846912906119584,
"eval_f1": 0.9091268664095719,
"eval_loss": 0.10913100093603134,
"eval_precision": 0.9004897159647405,
"eval_recall": 0.9179313099041534,
"eval_runtime": 25.3619,
"eval_samples_per_second": 128.145,
"eval_steps_per_second": 16.048,
"step": 14600
},
{
"epoch": 33.71,
"eval_accuracy_score": 0.9848278013963103,
"eval_f1": 0.9108891306497873,
"eval_loss": 0.10917215794324875,
"eval_precision": 0.9024103468547913,
"eval_recall": 0.9195287539936102,
"eval_runtime": 25.4358,
"eval_samples_per_second": 127.773,
"eval_steps_per_second": 16.001,
"step": 14800
},
{
"epoch": 34.17,
"learning_rate": 0.0,
"loss": 0.0003,
"step": 15000
},
{
"epoch": 34.17,
"eval_accuracy_score": 0.9847692967744451,
"eval_f1": 0.9102957175353575,
"eval_loss": 0.1090984046459198,
"eval_precision": 0.9018224573780129,
"eval_recall": 0.9189297124600639,
"eval_runtime": 25.2915,
"eval_samples_per_second": 128.501,
"eval_steps_per_second": 16.092,
"step": 15000
},
{
"epoch": 34.17,
"step": 15000,
"total_flos": 7.879039709550797e+16,
"train_loss": 0.023325540216763814,
"train_runtime": 9228.2252,
"train_samples_per_second": 52.014,
"train_steps_per_second": 1.625
}
],
"max_steps": 15000,
"num_train_epochs": 35,
"total_flos": 7.879039709550797e+16,
"trial_name": null,
"trial_params": null
}