qwen-compassion-web-s1-part1 / trainer_state.json
sparrow8i8's picture
add trainer_state.json
d735162 verified
Raw History Blame Contribute Delete
24.8 kB
{
"best_global_step": 600,
"best_metric": 0.26600781083106995,
"best_model_checkpoint": "/workspace/anchor/qwen-compassion-web-s1/train/checkpoint-600",
"epoch": 1.0,
"eval_steps": 50,
"global_step": 614,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.008143322475570033,
"grad_norm": 7.778277397155762,
"learning_rate": 2.5806451612903226e-05,
"loss": 1.2146,
"step": 5
},
{
"epoch": 0.016286644951140065,
"grad_norm": 3.7038655281066895,
"learning_rate": 5.8064516129032266e-05,
"loss": 0.9438,
"step": 10
},
{
"epoch": 0.024429967426710098,
"grad_norm": 2.6018831729888916,
"learning_rate": 9.032258064516129e-05,
"loss": 0.8658,
"step": 15
},
{
"epoch": 0.03257328990228013,
"grad_norm": 1.9596905708312988,
"learning_rate": 0.00012258064516129034,
"loss": 0.8249,
"step": 20
},
{
"epoch": 0.04071661237785016,
"grad_norm": 3.6029365062713623,
"learning_rate": 0.00015483870967741937,
"loss": 0.7622,
"step": 25
},
{
"epoch": 0.048859934853420196,
"grad_norm": 2.8365094661712646,
"learning_rate": 0.0001870967741935484,
"loss": 0.6029,
"step": 30
},
{
"epoch": 0.057003257328990226,
"grad_norm": 2.2653887271881104,
"learning_rate": 0.00019998693330688282,
"loss": 0.6964,
"step": 35
},
{
"epoch": 0.06514657980456026,
"grad_norm": 2.8496718406677246,
"learning_rate": 0.00019990709365929677,
"loss": 0.555,
"step": 40
},
{
"epoch": 0.0732899022801303,
"grad_norm": 2.1634292602539062,
"learning_rate": 0.00019975473152335726,
"loss": 0.5607,
"step": 45
},
{
"epoch": 0.08143322475570032,
"grad_norm": 1.7043299674987793,
"learning_rate": 0.0001995299574986306,
"loss": 0.5364,
"step": 50
},
{
"epoch": 0.08143322475570032,
"eval_loss": 0.3740517795085907,
"eval_runtime": 168.3163,
"eval_samples_per_second": 1.497,
"eval_steps_per_second": 0.749,
"step": 50
},
{
"epoch": 0.08957654723127036,
"grad_norm": 1.4339250326156616,
"learning_rate": 0.00019923293474842174,
"loss": 0.5874,
"step": 55
},
{
"epoch": 0.09771986970684039,
"grad_norm": 1.223159670829773,
"learning_rate": 0.00019886387888133413,
"loss": 0.491,
"step": 60
},
{
"epoch": 0.10586319218241043,
"grad_norm": 1.2165244817733765,
"learning_rate": 0.00019842305779475968,
"loss": 0.388,
"step": 65
},
{
"epoch": 0.11400651465798045,
"grad_norm": 1.350937843322754,
"learning_rate": 0.0001979107914804122,
"loss": 0.5532,
"step": 70
},
{
"epoch": 0.12214983713355049,
"grad_norm": 1.0507100820541382,
"learning_rate": 0.00019732745179204552,
"loss": 0.4037,
"step": 75
},
{
"epoch": 0.13029315960912052,
"grad_norm": 1.7351406812667847,
"learning_rate": 0.00019667346217552527,
"loss": 0.4045,
"step": 80
},
{
"epoch": 0.13843648208469056,
"grad_norm": 1.3686401844024658,
"learning_rate": 0.00019594929736144976,
"loss": 0.5843,
"step": 85
},
{
"epoch": 0.1465798045602606,
"grad_norm": 1.2211040258407593,
"learning_rate": 0.00019515548302054335,
"loss": 0.4779,
"step": 90
},
{
"epoch": 0.15472312703583063,
"grad_norm": 1.103126883506775,
"learning_rate": 0.0001942925953820725,
"loss": 0.4436,
"step": 95
},
{
"epoch": 0.16286644951140064,
"grad_norm": 1.138192057609558,
"learning_rate": 0.00019336126081556134,
"loss": 0.5071,
"step": 100
},
{
"epoch": 0.16286644951140064,
"eval_loss": 0.3115701377391815,
"eval_runtime": 82.361,
"eval_samples_per_second": 3.06,
"eval_steps_per_second": 1.53,
"step": 100
},
{
"epoch": 0.17100977198697068,
"grad_norm": 1.185229778289795,
"learning_rate": 0.00019236215537611046,
"loss": 0.4371,
"step": 105
},
{
"epoch": 0.1791530944625407,
"grad_norm": 1.2231813669204712,
"learning_rate": 0.00019129600431364897,
"loss": 0.4514,
"step": 110
},
{
"epoch": 0.18729641693811075,
"grad_norm": 1.2442817687988281,
"learning_rate": 0.00019016358154647618,
"loss": 0.446,
"step": 115
},
{
"epoch": 0.19543973941368079,
"grad_norm": 1.3569902181625366,
"learning_rate": 0.00018896570909947475,
"loss": 0.4254,
"step": 120
},
{
"epoch": 0.20358306188925082,
"grad_norm": 1.1408485174179077,
"learning_rate": 0.00018770325650740345,
"loss": 0.4192,
"step": 125
},
{
"epoch": 0.21172638436482086,
"grad_norm": 1.190841555595398,
"learning_rate": 0.00018637714018370253,
"loss": 0.4725,
"step": 130
},
{
"epoch": 0.21986970684039087,
"grad_norm": 0.8341662883758545,
"learning_rate": 0.00018498832275526988,
"loss": 0.4289,
"step": 135
},
{
"epoch": 0.2280130293159609,
"grad_norm": 0.6739444136619568,
"learning_rate": 0.00018353781236369064,
"loss": 0.4329,
"step": 140
},
{
"epoch": 0.23615635179153094,
"grad_norm": 1.1772111654281616,
"learning_rate": 0.00018202666193342833,
"loss": 0.4807,
"step": 145
},
{
"epoch": 0.24429967426710097,
"grad_norm": 1.3342751264572144,
"learning_rate": 0.00018045596840750723,
"loss": 0.4775,
"step": 150
},
{
"epoch": 0.24429967426710097,
"eval_loss": 0.29967936873435974,
"eval_runtime": 83.3439,
"eval_samples_per_second": 3.024,
"eval_steps_per_second": 1.512,
"step": 150
},
{
"epoch": 0.252442996742671,
"grad_norm": 0.9628245234489441,
"learning_rate": 0.0001788268719512427,
"loss": 0.4429,
"step": 155
},
{
"epoch": 0.26058631921824105,
"grad_norm": 1.3705332279205322,
"learning_rate": 0.00017714055512459565,
"loss": 0.4301,
"step": 160
},
{
"epoch": 0.2687296416938111,
"grad_norm": 0.9929981231689453,
"learning_rate": 0.0001753982420237533,
"loss": 0.5026,
"step": 165
},
{
"epoch": 0.2768729641693811,
"grad_norm": 1.2665433883666992,
"learning_rate": 0.00017360119739255852,
"loss": 0.3337,
"step": 170
},
{
"epoch": 0.28501628664495116,
"grad_norm": 0.7617188096046448,
"learning_rate": 0.00017175072570443312,
"loss": 0.4805,
"step": 175
},
{
"epoch": 0.2931596091205212,
"grad_norm": 0.868850588798523,
"learning_rate": 0.00016984817021546177,
"loss": 0.3722,
"step": 180
},
{
"epoch": 0.30130293159609123,
"grad_norm": 1.6269251108169556,
"learning_rate": 0.000167894911989323,
"loss": 0.426,
"step": 185
},
{
"epoch": 0.30944625407166126,
"grad_norm": 1.1786128282546997,
"learning_rate": 0.00016589236889477646,
"loss": 0.3369,
"step": 190
},
{
"epoch": 0.31758957654723124,
"grad_norm": 1.209784984588623,
"learning_rate": 0.00016384199457643262,
"loss": 0.4643,
"step": 195
},
{
"epoch": 0.3257328990228013,
"grad_norm": 1.0796294212341309,
"learning_rate": 0.00016174527739955342,
"loss": 0.5805,
"step": 200
},
{
"epoch": 0.3257328990228013,
"eval_loss": 0.2914806008338928,
"eval_runtime": 83.421,
"eval_samples_per_second": 3.021,
"eval_steps_per_second": 1.51,
"step": 200
},
{
"epoch": 0.3338762214983713,
"grad_norm": 0.6735913753509521,
"learning_rate": 0.00015960373936964892,
"loss": 0.3844,
"step": 205
},
{
"epoch": 0.34201954397394135,
"grad_norm": 1.1781941652297974,
"learning_rate": 0.0001574189350276545,
"loss": 0.4737,
"step": 210
},
{
"epoch": 0.3501628664495114,
"grad_norm": 1.2786900997161865,
"learning_rate": 0.00015519245032149083,
"loss": 0.52,
"step": 215
},
{
"epoch": 0.3583061889250814,
"grad_norm": 0.9720235466957092,
"learning_rate": 0.0001529259014548253,
"loss": 0.4297,
"step": 220
},
{
"epoch": 0.36644951140065146,
"grad_norm": 1.1906429529190063,
"learning_rate": 0.00015062093371387097,
"loss": 0.4357,
"step": 225
},
{
"epoch": 0.3745928338762215,
"grad_norm": 1.2936455011367798,
"learning_rate": 0.00014827922027307451,
"loss": 0.403,
"step": 230
},
{
"epoch": 0.38273615635179153,
"grad_norm": 1.0195300579071045,
"learning_rate": 0.00014590246098055996,
"loss": 0.4624,
"step": 235
},
{
"epoch": 0.39087947882736157,
"grad_norm": 0.8488811254501343,
"learning_rate": 0.00014349238112421024,
"loss": 0.3107,
"step": 240
},
{
"epoch": 0.3990228013029316,
"grad_norm": 0.957468569278717,
"learning_rate": 0.000141050730179282,
"loss": 0.4091,
"step": 245
},
{
"epoch": 0.40716612377850164,
"grad_norm": 1.1673383712768555,
"learning_rate": 0.0001385792805384625,
"loss": 0.5034,
"step": 250
},
{
"epoch": 0.40716612377850164,
"eval_loss": 0.2860499620437622,
"eval_runtime": 83.1347,
"eval_samples_per_second": 3.031,
"eval_steps_per_second": 1.516,
"step": 250
},
{
"epoch": 0.4153094462540717,
"grad_norm": 1.116147756576538,
"learning_rate": 0.00013607982622529133,
"loss": 0.4086,
"step": 255
},
{
"epoch": 0.4234527687296417,
"grad_norm": 1.2793134450912476,
"learning_rate": 0.00013355418159187985,
"loss": 0.4773,
"step": 260
},
{
"epoch": 0.43159609120521175,
"grad_norm": 0.6993953585624695,
"learning_rate": 0.00013100418000187419,
"loss": 0.4601,
"step": 265
},
{
"epoch": 0.43973941368078173,
"grad_norm": 1.048805594444275,
"learning_rate": 0.0001284316724996181,
"loss": 0.459,
"step": 270
},
{
"epoch": 0.44788273615635177,
"grad_norm": 1.164232850074768,
"learning_rate": 0.00012583852646648095,
"loss": 0.5734,
"step": 275
},
{
"epoch": 0.4560260586319218,
"grad_norm": 1.1234965324401855,
"learning_rate": 0.00012322662426532708,
"loss": 0.4605,
"step": 280
},
{
"epoch": 0.46416938110749184,
"grad_norm": 0.6971266865730286,
"learning_rate": 0.00012059786187410984,
"loss": 0.3028,
"step": 285
},
{
"epoch": 0.4723127035830619,
"grad_norm": 0.6424208879470825,
"learning_rate": 0.00011795414750958265,
"loss": 0.2779,
"step": 290
},
{
"epoch": 0.4804560260586319,
"grad_norm": 1.0949883460998535,
"learning_rate": 0.00011529740024212565,
"loss": 0.4043,
"step": 295
},
{
"epoch": 0.48859934853420195,
"grad_norm": 1.2397624254226685,
"learning_rate": 0.00011262954860269399,
"loss": 0.5491,
"step": 300
},
{
"epoch": 0.48859934853420195,
"eval_loss": 0.2816900312900543,
"eval_runtime": 83.2171,
"eval_samples_per_second": 3.028,
"eval_steps_per_second": 1.514,
"step": 300
},
{
"epoch": 0.496742671009772,
"grad_norm": 1.1865627765655518,
"learning_rate": 0.0001099525291828986,
"loss": 0.3669,
"step": 305
},
{
"epoch": 0.504885993485342,
"grad_norm": 1.2804713249206543,
"learning_rate": 0.00010726828522923562,
"loss": 0.4,
"step": 310
},
{
"epoch": 0.5130293159609121,
"grad_norm": 1.200066328048706,
"learning_rate": 0.00010457876523248518,
"loss": 0.4635,
"step": 315
},
{
"epoch": 0.5211726384364821,
"grad_norm": 0.629518985748291,
"learning_rate": 0.00010188592151330343,
"loss": 0.3406,
"step": 320
},
{
"epoch": 0.5293159609120521,
"grad_norm": 0.6699548363685608,
"learning_rate": 9.919170880503415e-05,
"loss": 0.409,
"step": 325
},
{
"epoch": 0.5374592833876222,
"grad_norm": 1.1119918823242188,
"learning_rate": 9.649808283476941e-05,
"loss": 0.2435,
"step": 330
},
{
"epoch": 0.5456026058631922,
"grad_norm": 0.7296743988990784,
"learning_rate": 9.38069989036886e-05,
"loss": 0.4156,
"step": 335
},
{
"epoch": 0.5537459283387622,
"grad_norm": 0.6966118216514587,
"learning_rate": 9.112041046770653e-05,
"loss": 0.2242,
"step": 340
},
{
"epoch": 0.5618892508143323,
"grad_norm": 1.0954296588897705,
"learning_rate": 8.844026771946147e-05,
"loss": 0.4235,
"step": 345
},
{
"epoch": 0.5700325732899023,
"grad_norm": 1.2122018337249756,
"learning_rate": 8.57685161726715e-05,
"loss": 0.4923,
"step": 350
},
{
"epoch": 0.5700325732899023,
"eval_loss": 0.2758687138557434,
"eval_runtime": 82.6942,
"eval_samples_per_second": 3.047,
"eval_steps_per_second": 1.524,
"step": 350
},
{
"epoch": 0.5781758957654723,
"grad_norm": 0.9051918983459473,
"learning_rate": 8.310709524988743e-05,
"loss": 0.433,
"step": 355
},
{
"epoch": 0.5863192182410424,
"grad_norm": 1.0195807218551636,
"learning_rate": 8.045793687466757e-05,
"loss": 0.3257,
"step": 360
},
{
"epoch": 0.5944625407166124,
"grad_norm": 0.9997521042823792,
"learning_rate": 7.782296406919557e-05,
"loss": 0.4662,
"step": 365
},
{
"epoch": 0.6026058631921825,
"grad_norm": 1.0739184617996216,
"learning_rate": 7.520408955836007e-05,
"loss": 0.2983,
"step": 370
},
{
"epoch": 0.6107491856677525,
"grad_norm": 0.6302400827407837,
"learning_rate": 7.260321438130913e-05,
"loss": 0.3302,
"step": 375
},
{
"epoch": 0.6188925081433225,
"grad_norm": 0.8717932105064392,
"learning_rate": 7.002222651148714e-05,
"loss": 0.2522,
"step": 380
},
{
"epoch": 0.6270358306188925,
"grad_norm": 1.3393378257751465,
"learning_rate": 6.746299948615631e-05,
"loss": 0.5712,
"step": 385
},
{
"epoch": 0.6351791530944625,
"grad_norm": 0.9017629623413086,
"learning_rate": 6.492739104639727e-05,
"loss": 0.3188,
"step": 390
},
{
"epoch": 0.6433224755700325,
"grad_norm": 0.8153979778289795,
"learning_rate": 6.24172417885762e-05,
"loss": 0.3989,
"step": 395
},
{
"epoch": 0.6514657980456026,
"grad_norm": 1.067793607711792,
"learning_rate": 5.9934373828257106e-05,
"loss": 0.3344,
"step": 400
},
{
"epoch": 0.6514657980456026,
"eval_loss": 0.2729613184928894,
"eval_runtime": 82.733,
"eval_samples_per_second": 3.046,
"eval_steps_per_second": 1.523,
"step": 400
},
{
"epoch": 0.6596091205211726,
"grad_norm": 0.6235025525093079,
"learning_rate": 5.7480589477529545e-05,
"loss": 0.3693,
"step": 405
},
{
"epoch": 0.6677524429967426,
"grad_norm": 0.9729871153831482,
"learning_rate": 5.505766993671162e-05,
"loss": 0.4579,
"step": 410
},
{
"epoch": 0.6758957654723127,
"grad_norm": 1.1238763332366943,
"learning_rate": 5.266737400137765e-05,
"loss": 0.2974,
"step": 415
},
{
"epoch": 0.6840390879478827,
"grad_norm": 1.3028978109359741,
"learning_rate": 5.031143678565005e-05,
"loss": 0.3374,
"step": 420
},
{
"epoch": 0.6921824104234527,
"grad_norm": 0.5976630449295044,
"learning_rate": 4.799156846268095e-05,
"loss": 0.3009,
"step": 425
},
{
"epoch": 0.7003257328990228,
"grad_norm": 1.2614715099334717,
"learning_rate": 4.57094530232389e-05,
"loss": 0.4365,
"step": 430
},
{
"epoch": 0.7084690553745928,
"grad_norm": 0.9816355109214783,
"learning_rate": 4.346674705330117e-05,
"loss": 0.4431,
"step": 435
},
{
"epoch": 0.7166123778501629,
"grad_norm": 1.0284030437469482,
"learning_rate": 4.1265078531538916e-05,
"loss": 0.3474,
"step": 440
},
{
"epoch": 0.7247557003257329,
"grad_norm": 1.1783193349838257,
"learning_rate": 3.9106045647569e-05,
"loss": 0.4137,
"step": 445
},
{
"epoch": 0.7328990228013029,
"grad_norm": 1.3315753936767578,
"learning_rate": 3.69912156418289e-05,
"loss": 0.5189,
"step": 450
},
{
"epoch": 0.7328990228013029,
"eval_loss": 0.2701830565929413,
"eval_runtime": 83.1814,
"eval_samples_per_second": 3.03,
"eval_steps_per_second": 1.515,
"step": 450
},
{
"epoch": 0.741042345276873,
"grad_norm": 1.1194369792938232,
"learning_rate": 3.492212366791831e-05,
"loss": 0.3809,
"step": 455
},
{
"epoch": 0.749185667752443,
"grad_norm": 1.0723215341567993,
"learning_rate": 3.290027167823204e-05,
"loss": 0.3827,
"step": 460
},
{
"epoch": 0.757328990228013,
"grad_norm": 1.043360948562622,
"learning_rate": 3.092712733369387e-05,
"loss": 0.3443,
"step": 465
},
{
"epoch": 0.7654723127035831,
"grad_norm": 0.7220495343208313,
"learning_rate": 2.9004122938382617e-05,
"loss": 0.3222,
"step": 470
},
{
"epoch": 0.7736156351791531,
"grad_norm": 0.9931464195251465,
"learning_rate": 2.7132654399823444e-05,
"loss": 0.428,
"step": 475
},
{
"epoch": 0.7817589576547231,
"grad_norm": 0.8456658124923706,
"learning_rate": 2.531408021569982e-05,
"loss": 0.3892,
"step": 480
},
{
"epoch": 0.7899022801302932,
"grad_norm": 0.9457429647445679,
"learning_rate": 2.3549720487720738e-05,
"loss": 0.4835,
"step": 485
},
{
"epoch": 0.7980456026058632,
"grad_norm": 1.0641571283340454,
"learning_rate": 2.184085596336011e-05,
"loss": 0.4096,
"step": 490
},
{
"epoch": 0.8061889250814332,
"grad_norm": 1.1258314847946167,
"learning_rate": 2.0188727106162874e-05,
"loss": 0.4558,
"step": 495
},
{
"epoch": 0.8143322475570033,
"grad_norm": 0.8964729905128479,
"learning_rate": 1.8594533195293427e-05,
"loss": 0.412,
"step": 500
},
{
"epoch": 0.8143322475570033,
"eval_loss": 0.26734381914138794,
"eval_runtime": 83.1973,
"eval_samples_per_second": 3.029,
"eval_steps_per_second": 1.514,
"step": 500
},
{
"epoch": 0.8224755700325733,
"grad_norm": 1.0101343393325806,
"learning_rate": 1.7059431454979824e-05,
"loss": 0.4203,
"step": 505
},
{
"epoch": 0.8306188925081434,
"grad_norm": 1.4090873003005981,
"learning_rate": 1.5584536214485457e-05,
"loss": 0.3865,
"step": 510
},
{
"epoch": 0.8387622149837134,
"grad_norm": 1.0419752597808838,
"learning_rate": 1.4170918099218166e-05,
"loss": 0.3799,
"step": 515
},
{
"epoch": 0.8469055374592834,
"grad_norm": 0.9638477563858032,
"learning_rate": 1.2819603253564205e-05,
"loss": 0.3535,
"step": 520
},
{
"epoch": 0.8550488599348535,
"grad_norm": 1.0044305324554443,
"learning_rate": 1.153157259601062e-05,
"loss": 0.3689,
"step": 525
},
{
"epoch": 0.8631921824104235,
"grad_norm": 0.9920660853385925,
"learning_rate": 1.030776110709718e-05,
"loss": 0.436,
"step": 530
},
{
"epoch": 0.8713355048859935,
"grad_norm": 1.024511456489563,
"learning_rate": 9.149057150714801e-06,
"loss": 0.4191,
"step": 535
},
{
"epoch": 0.8794788273615635,
"grad_norm": 1.0346133708953857,
"learning_rate": 8.056301829242784e-06,
"loss": 0.3713,
"step": 540
},
{
"epoch": 0.8876221498371335,
"grad_norm": 0.6623178124427795,
"learning_rate": 7.030288372993066e-06,
"loss": 0.2473,
"step": 545
},
{
"epoch": 0.8957654723127035,
"grad_norm": 1.0537056922912598,
"learning_rate": 6.0717615644051206e-06,
"loss": 0.4447,
"step": 550
},
{
"epoch": 0.8957654723127035,
"eval_loss": 0.2662525773048401,
"eval_runtime": 82.6125,
"eval_samples_per_second": 3.05,
"eval_steps_per_second": 1.525,
"step": 550
},
{
"epoch": 0.9039087947882736,
"grad_norm": 0.8485147953033447,
"learning_rate": 5.181417197408734e-06,
"loss": 0.4455,
"step": 555
},
{
"epoch": 0.9120521172638436,
"grad_norm": 0.8799261450767517,
"learning_rate": 4.359901572347758e-06,
"loss": 0.3891,
"step": 560
},
{
"epoch": 0.9201954397394136,
"grad_norm": 1.0102251768112183,
"learning_rate": 3.607811026831176e-06,
"loss": 0.4678,
"step": 565
},
{
"epoch": 0.9283387622149837,
"grad_norm": 1.056903600692749,
"learning_rate": 2.9256915028519573e-06,
"loss": 0.3148,
"step": 570
},
{
"epoch": 0.9364820846905537,
"grad_norm": 0.9595625996589661,
"learning_rate": 2.3140381504882737e-06,
"loss": 0.4204,
"step": 575
},
{
"epoch": 0.9446254071661238,
"grad_norm": 0.906981885433197,
"learning_rate": 1.7732949684743594e-06,
"loss": 0.3291,
"step": 580
},
{
"epoch": 0.9527687296416938,
"grad_norm": 1.2913422584533691,
"learning_rate": 1.303854481902067e-06,
"loss": 0.3913,
"step": 585
},
{
"epoch": 0.9609120521172638,
"grad_norm": 1.0793213844299316,
"learning_rate": 9.060574572873237e-07,
"loss": 0.3904,
"step": 590
},
{
"epoch": 0.9690553745928339,
"grad_norm": 1.1022560596466064,
"learning_rate": 5.801926552078563e-07,
"loss": 0.3607,
"step": 595
},
{
"epoch": 0.9771986970684039,
"grad_norm": 1.1481983661651611,
"learning_rate": 3.264966206921294e-07,
"loss": 0.3837,
"step": 600
},
{
"epoch": 0.9771986970684039,
"eval_loss": 0.26600781083106995,
"eval_runtime": 83.4266,
"eval_samples_per_second": 3.021,
"eval_steps_per_second": 1.51,
"step": 600
},
{
"epoch": 0.9853420195439739,
"grad_norm": 1.2523059844970703,
"learning_rate": 1.451535115114866e-07,
"loss": 0.4497,
"step": 605
},
{
"epoch": 0.993485342019544,
"grad_norm": 0.9074209332466125,
"learning_rate": 3.629496450011649e-08,
"loss": 0.3181,
"step": 610
}
],
"logging_steps": 5,
"max_steps": 614,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 2,
"early_stopping_threshold": 0.001
},
"attributes": {
"early_stopping_patience_counter": 1
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2.5080656292484915e+18,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}