| { | |
| "best_metric": 0.9360692350642099, | |
| "best_model_checkpoint": "/tmp/classification_phobertlarge/checkpoint-6198", | |
| "epoch": 40.0, | |
| "eval_steps": 500, | |
| "global_step": 41320, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.484027105517909, | |
| "grad_norm": 107.85002136230469, | |
| "learning_rate": 2.9636979670861567e-05, | |
| "loss": 0.4143, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.968054211035818, | |
| "grad_norm": 15.951617240905762, | |
| "learning_rate": 2.927395934172314e-05, | |
| "loss": 0.3094, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.9327191513121161, | |
| "eval_loss": 0.22949737310409546, | |
| "eval_runtime": 30.2358, | |
| "eval_samples_per_second": 118.469, | |
| "eval_steps_per_second": 4.961, | |
| "step": 1033 | |
| }, | |
| { | |
| "epoch": 1.452081316553727, | |
| "grad_norm": 0.18993481993675232, | |
| "learning_rate": 2.8910939012584705e-05, | |
| "loss": 0.2335, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.936108422071636, | |
| "grad_norm": 2.2120251655578613, | |
| "learning_rate": 2.8547918683446275e-05, | |
| "loss": 0.2346, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.9279731993299832, | |
| "eval_loss": 0.2650923430919647, | |
| "eval_runtime": 30.2762, | |
| "eval_samples_per_second": 118.311, | |
| "eval_steps_per_second": 4.954, | |
| "step": 2066 | |
| }, | |
| { | |
| "epoch": 2.420135527589545, | |
| "grad_norm": 1.64552640914917, | |
| "learning_rate": 2.818489835430784e-05, | |
| "loss": 0.1957, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.904162633107454, | |
| "grad_norm": 1.3047701120376587, | |
| "learning_rate": 2.7821878025169413e-05, | |
| "loss": 0.2359, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.9338358458961474, | |
| "eval_loss": 0.2357342690229416, | |
| "eval_runtime": 30.2251, | |
| "eval_samples_per_second": 118.511, | |
| "eval_steps_per_second": 4.963, | |
| "step": 3099 | |
| }, | |
| { | |
| "epoch": 3.388189738625363, | |
| "grad_norm": 21.052520751953125, | |
| "learning_rate": 2.745885769603098e-05, | |
| "loss": 0.179, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.872216844143272, | |
| "grad_norm": 0.44238439202308655, | |
| "learning_rate": 2.7095837366892545e-05, | |
| "loss": 0.1734, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_f1": 0.9276940256839754, | |
| "eval_loss": 0.2720405161380768, | |
| "eval_runtime": 30.1857, | |
| "eval_samples_per_second": 118.666, | |
| "eval_steps_per_second": 4.969, | |
| "step": 4132 | |
| }, | |
| { | |
| "epoch": 4.356243949661181, | |
| "grad_norm": 0.10865873098373413, | |
| "learning_rate": 2.6732817037754114e-05, | |
| "loss": 0.1593, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.84027105517909, | |
| "grad_norm": 35.33164978027344, | |
| "learning_rate": 2.6369796708615684e-05, | |
| "loss": 0.1551, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_f1": 0.9349525404801786, | |
| "eval_loss": 0.29397937655448914, | |
| "eval_runtime": 30.2085, | |
| "eval_samples_per_second": 118.576, | |
| "eval_steps_per_second": 4.965, | |
| "step": 5165 | |
| }, | |
| { | |
| "epoch": 5.3242981606969995, | |
| "grad_norm": 0.6579780578613281, | |
| "learning_rate": 2.6006776379477253e-05, | |
| "loss": 0.1434, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.808325266214908, | |
| "grad_norm": 0.2094825804233551, | |
| "learning_rate": 2.564375605033882e-05, | |
| "loss": 0.1282, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_f1": 0.9360692350642099, | |
| "eval_loss": 0.2751626968383789, | |
| "eval_runtime": 30.1054, | |
| "eval_samples_per_second": 118.982, | |
| "eval_steps_per_second": 4.983, | |
| "step": 6198 | |
| }, | |
| { | |
| "epoch": 6.292352371732817, | |
| "grad_norm": 0.12414942681789398, | |
| "learning_rate": 2.5280735721200388e-05, | |
| "loss": 0.1158, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 6.776379477250726, | |
| "grad_norm": 4.030830383300781, | |
| "learning_rate": 2.4917715392061957e-05, | |
| "loss": 0.1208, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_f1": 0.9346733668341709, | |
| "eval_loss": 0.3136898875236511, | |
| "eval_runtime": 30.2127, | |
| "eval_samples_per_second": 118.559, | |
| "eval_steps_per_second": 4.965, | |
| "step": 7231 | |
| }, | |
| { | |
| "epoch": 7.260406582768635, | |
| "grad_norm": 4.0914530754089355, | |
| "learning_rate": 2.4554695062923523e-05, | |
| "loss": 0.1127, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 7.744433688286544, | |
| "grad_norm": 0.10396777093410492, | |
| "learning_rate": 2.4191674733785093e-05, | |
| "loss": 0.1074, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_f1": 0.9355108877721943, | |
| "eval_loss": 0.2966548800468445, | |
| "eval_runtime": 30.2555, | |
| "eval_samples_per_second": 118.392, | |
| "eval_steps_per_second": 4.958, | |
| "step": 8264 | |
| }, | |
| { | |
| "epoch": 8.228460793804453, | |
| "grad_norm": 0.22765442728996277, | |
| "learning_rate": 2.382865440464666e-05, | |
| "loss": 0.0993, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 8.712487899322362, | |
| "grad_norm": 0.5725598931312561, | |
| "learning_rate": 2.346563407550823e-05, | |
| "loss": 0.2719, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8269146084785461, | |
| "eval_runtime": 30.1812, | |
| "eval_samples_per_second": 118.683, | |
| "eval_steps_per_second": 4.97, | |
| "step": 9297 | |
| }, | |
| { | |
| "epoch": 9.196515004840272, | |
| "grad_norm": 1.2485491037368774, | |
| "learning_rate": 2.3102613746369797e-05, | |
| "loss": 0.8488, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 9.68054211035818, | |
| "grad_norm": 0.6287474036216736, | |
| "learning_rate": 2.2739593417231366e-05, | |
| "loss": 0.8456, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8219632506370544, | |
| "eval_runtime": 30.1694, | |
| "eval_samples_per_second": 118.729, | |
| "eval_steps_per_second": 4.972, | |
| "step": 10330 | |
| }, | |
| { | |
| "epoch": 10.164569215876089, | |
| "grad_norm": 1.4201635122299194, | |
| "learning_rate": 2.2376573088092936e-05, | |
| "loss": 0.8446, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 10.648596321393999, | |
| "grad_norm": 1.115445613861084, | |
| "learning_rate": 2.20135527589545e-05, | |
| "loss": 0.84, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8211880922317505, | |
| "eval_runtime": 30.1617, | |
| "eval_samples_per_second": 118.76, | |
| "eval_steps_per_second": 4.973, | |
| "step": 11363 | |
| }, | |
| { | |
| "epoch": 11.132623426911907, | |
| "grad_norm": 1.784400463104248, | |
| "learning_rate": 2.165053242981607e-05, | |
| "loss": 0.8435, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 11.616650532429816, | |
| "grad_norm": 1.7544752359390259, | |
| "learning_rate": 2.1287512100677637e-05, | |
| "loss": 0.841, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8207370042800903, | |
| "eval_runtime": 30.1494, | |
| "eval_samples_per_second": 118.808, | |
| "eval_steps_per_second": 4.975, | |
| "step": 12396 | |
| }, | |
| { | |
| "epoch": 12.100677637947726, | |
| "grad_norm": 0.660882830619812, | |
| "learning_rate": 2.092449177153921e-05, | |
| "loss": 0.8421, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 12.584704743465634, | |
| "grad_norm": 0.5564029812812805, | |
| "learning_rate": 2.0561471442400775e-05, | |
| "loss": 0.8383, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8198010325431824, | |
| "eval_runtime": 30.2024, | |
| "eval_samples_per_second": 118.6, | |
| "eval_steps_per_second": 4.966, | |
| "step": 13429 | |
| }, | |
| { | |
| "epoch": 13.068731848983543, | |
| "grad_norm": 0.45510706305503845, | |
| "learning_rate": 2.019845111326234e-05, | |
| "loss": 0.8494, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 13.552758954501453, | |
| "grad_norm": 0.9993030428886414, | |
| "learning_rate": 1.983543078412391e-05, | |
| "loss": 0.8371, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8276057839393616, | |
| "eval_runtime": 30.1498, | |
| "eval_samples_per_second": 118.807, | |
| "eval_steps_per_second": 4.975, | |
| "step": 14462 | |
| }, | |
| { | |
| "epoch": 14.036786060019361, | |
| "grad_norm": 0.8151612281799316, | |
| "learning_rate": 1.947241045498548e-05, | |
| "loss": 0.8465, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 14.52081316553727, | |
| "grad_norm": 1.312504768371582, | |
| "learning_rate": 1.910939012584705e-05, | |
| "loss": 0.8486, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8243775367736816, | |
| "eval_runtime": 30.177, | |
| "eval_samples_per_second": 118.7, | |
| "eval_steps_per_second": 4.971, | |
| "step": 15495 | |
| }, | |
| { | |
| "epoch": 15.00484027105518, | |
| "grad_norm": 1.09600031375885, | |
| "learning_rate": 1.8746369796708615e-05, | |
| "loss": 0.8345, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 15.488867376573088, | |
| "grad_norm": 1.3698476552963257, | |
| "learning_rate": 1.8383349467570184e-05, | |
| "loss": 0.8365, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 15.972894482090997, | |
| "grad_norm": 0.5735417008399963, | |
| "learning_rate": 1.8020329138431754e-05, | |
| "loss": 0.844, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8390949964523315, | |
| "eval_runtime": 30.2033, | |
| "eval_samples_per_second": 118.596, | |
| "eval_steps_per_second": 4.966, | |
| "step": 16528 | |
| }, | |
| { | |
| "epoch": 16.456921587608907, | |
| "grad_norm": 1.3362003564834595, | |
| "learning_rate": 1.765730880929332e-05, | |
| "loss": 0.8349, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 16.940948693126813, | |
| "grad_norm": 0.4842800796031952, | |
| "learning_rate": 1.729428848015489e-05, | |
| "loss": 0.837, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8234853744506836, | |
| "eval_runtime": 30.2033, | |
| "eval_samples_per_second": 118.596, | |
| "eval_steps_per_second": 4.966, | |
| "step": 17561 | |
| }, | |
| { | |
| "epoch": 17.424975798644724, | |
| "grad_norm": 0.564506471157074, | |
| "learning_rate": 1.6931268151016455e-05, | |
| "loss": 0.8368, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 17.909002904162634, | |
| "grad_norm": 1.0698164701461792, | |
| "learning_rate": 1.6568247821878027e-05, | |
| "loss": 0.8438, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8246738314628601, | |
| "eval_runtime": 30.1891, | |
| "eval_samples_per_second": 118.652, | |
| "eval_steps_per_second": 4.969, | |
| "step": 18594 | |
| }, | |
| { | |
| "epoch": 18.393030009680544, | |
| "grad_norm": 0.8105427026748657, | |
| "learning_rate": 1.6205227492739593e-05, | |
| "loss": 0.8367, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 18.87705711519845, | |
| "grad_norm": 0.630368173122406, | |
| "learning_rate": 1.5842207163601163e-05, | |
| "loss": 0.8418, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8236768841743469, | |
| "eval_runtime": 30.1711, | |
| "eval_samples_per_second": 118.723, | |
| "eval_steps_per_second": 4.972, | |
| "step": 19627 | |
| }, | |
| { | |
| "epoch": 19.36108422071636, | |
| "grad_norm": 0.884172797203064, | |
| "learning_rate": 1.5479186834462732e-05, | |
| "loss": 0.8503, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 19.845111326234267, | |
| "grad_norm": 0.585003674030304, | |
| "learning_rate": 1.51161665053243e-05, | |
| "loss": 0.8384, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8199198246002197, | |
| "eval_runtime": 30.1335, | |
| "eval_samples_per_second": 118.871, | |
| "eval_steps_per_second": 4.978, | |
| "step": 20660 | |
| }, | |
| { | |
| "epoch": 20.329138431752177, | |
| "grad_norm": 1.194004774093628, | |
| "learning_rate": 1.4753146176185867e-05, | |
| "loss": 0.845, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 20.813165537270088, | |
| "grad_norm": 0.378979355096817, | |
| "learning_rate": 1.4390125847047436e-05, | |
| "loss": 0.8387, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8225679993629456, | |
| "eval_runtime": 29.9919, | |
| "eval_samples_per_second": 119.432, | |
| "eval_steps_per_second": 5.001, | |
| "step": 21693 | |
| }, | |
| { | |
| "epoch": 21.297192642787998, | |
| "grad_norm": 0.6615312099456787, | |
| "learning_rate": 1.4027105517909002e-05, | |
| "loss": 0.833, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 21.781219748305904, | |
| "grad_norm": 0.4334174394607544, | |
| "learning_rate": 1.3664085188770572e-05, | |
| "loss": 0.8478, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8205086588859558, | |
| "eval_runtime": 29.9624, | |
| "eval_samples_per_second": 119.55, | |
| "eval_steps_per_second": 5.006, | |
| "step": 22726 | |
| }, | |
| { | |
| "epoch": 22.265246853823815, | |
| "grad_norm": 0.8907693028450012, | |
| "learning_rate": 1.3301064859632139e-05, | |
| "loss": 0.8442, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 22.749273959341725, | |
| "grad_norm": 0.5629915595054626, | |
| "learning_rate": 1.2938044530493708e-05, | |
| "loss": 0.8364, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8259473443031311, | |
| "eval_runtime": 29.9524, | |
| "eval_samples_per_second": 119.59, | |
| "eval_steps_per_second": 5.008, | |
| "step": 23759 | |
| }, | |
| { | |
| "epoch": 23.23330106485963, | |
| "grad_norm": 0.47178810834884644, | |
| "learning_rate": 1.2575024201355276e-05, | |
| "loss": 0.83, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 23.71732817037754, | |
| "grad_norm": 0.5889131426811218, | |
| "learning_rate": 1.2212003872216845e-05, | |
| "loss": 0.8325, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8245280385017395, | |
| "eval_runtime": 29.9556, | |
| "eval_samples_per_second": 119.577, | |
| "eval_steps_per_second": 5.007, | |
| "step": 24792 | |
| }, | |
| { | |
| "epoch": 24.20135527589545, | |
| "grad_norm": 0.8042486310005188, | |
| "learning_rate": 1.1848983543078413e-05, | |
| "loss": 0.8357, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 24.68538238141336, | |
| "grad_norm": 0.7427254319190979, | |
| "learning_rate": 1.148596321393998e-05, | |
| "loss": 0.8289, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8248125314712524, | |
| "eval_runtime": 30.0337, | |
| "eval_samples_per_second": 119.266, | |
| "eval_steps_per_second": 4.994, | |
| "step": 25825 | |
| }, | |
| { | |
| "epoch": 25.16940948693127, | |
| "grad_norm": 1.4444160461425781, | |
| "learning_rate": 1.1122942884801548e-05, | |
| "loss": 0.8344, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 25.65343659244918, | |
| "grad_norm": 1.6647661924362183, | |
| "learning_rate": 1.0759922555663117e-05, | |
| "loss": 0.8251, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8246968984603882, | |
| "eval_runtime": 30.0343, | |
| "eval_samples_per_second": 119.263, | |
| "eval_steps_per_second": 4.994, | |
| "step": 26858 | |
| }, | |
| { | |
| "epoch": 26.137463697967085, | |
| "grad_norm": 0.560504138469696, | |
| "learning_rate": 1.0396902226524685e-05, | |
| "loss": 0.8265, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 26.621490803484996, | |
| "grad_norm": 2.6784770488739014, | |
| "learning_rate": 1.0033881897386254e-05, | |
| "loss": 0.824, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.821439802646637, | |
| "eval_runtime": 29.9327, | |
| "eval_samples_per_second": 119.668, | |
| "eval_steps_per_second": 5.011, | |
| "step": 27891 | |
| }, | |
| { | |
| "epoch": 27.105517909002906, | |
| "grad_norm": 0.9694743156433105, | |
| "learning_rate": 9.670861568247822e-06, | |
| "loss": 0.8304, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 27.589545014520812, | |
| "grad_norm": 0.8298953771591187, | |
| "learning_rate": 9.307841239109391e-06, | |
| "loss": 0.8197, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8281795382499695, | |
| "eval_runtime": 29.9052, | |
| "eval_samples_per_second": 119.778, | |
| "eval_steps_per_second": 5.016, | |
| "step": 28924 | |
| }, | |
| { | |
| "epoch": 28.073572120038722, | |
| "grad_norm": 1.530045986175537, | |
| "learning_rate": 8.944820909970959e-06, | |
| "loss": 0.8214, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 28.557599225556633, | |
| "grad_norm": 1.6190696954727173, | |
| "learning_rate": 8.581800580832526e-06, | |
| "loss": 0.8241, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8340330123901367, | |
| "eval_runtime": 29.9993, | |
| "eval_samples_per_second": 119.403, | |
| "eval_steps_per_second": 5.0, | |
| "step": 29957 | |
| }, | |
| { | |
| "epoch": 29.04162633107454, | |
| "grad_norm": 0.6487633585929871, | |
| "learning_rate": 8.218780251694096e-06, | |
| "loss": 0.8209, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 29.52565343659245, | |
| "grad_norm": 1.1463664770126343, | |
| "learning_rate": 7.855759922555663e-06, | |
| "loss": 0.8285, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8359894156455994, | |
| "eval_runtime": 29.9709, | |
| "eval_samples_per_second": 119.516, | |
| "eval_steps_per_second": 5.005, | |
| "step": 30990 | |
| }, | |
| { | |
| "epoch": 30.00968054211036, | |
| "grad_norm": 1.2427114248275757, | |
| "learning_rate": 7.492739593417232e-06, | |
| "loss": 0.8103, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 30.493707647628266, | |
| "grad_norm": 1.2282146215438843, | |
| "learning_rate": 7.129719264278799e-06, | |
| "loss": 0.814, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 30.977734753146176, | |
| "grad_norm": 1.6168450117111206, | |
| "learning_rate": 6.766698935140368e-06, | |
| "loss": 0.8169, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8400696516036987, | |
| "eval_runtime": 29.9824, | |
| "eval_samples_per_second": 119.47, | |
| "eval_steps_per_second": 5.003, | |
| "step": 32023 | |
| }, | |
| { | |
| "epoch": 31.461761858664087, | |
| "grad_norm": 2.838101863861084, | |
| "learning_rate": 6.403678606001936e-06, | |
| "loss": 0.8113, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 31.945788964181993, | |
| "grad_norm": 1.7709537744522095, | |
| "learning_rate": 6.040658276863505e-06, | |
| "loss": 0.811, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.853431224822998, | |
| "eval_runtime": 30.0224, | |
| "eval_samples_per_second": 119.311, | |
| "eval_steps_per_second": 4.996, | |
| "step": 33056 | |
| }, | |
| { | |
| "epoch": 32.4298160696999, | |
| "grad_norm": 1.901810646057129, | |
| "learning_rate": 5.677637947725073e-06, | |
| "loss": 0.8153, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 32.91384317521781, | |
| "grad_norm": 3.1245410442352295, | |
| "learning_rate": 5.3146176185866415e-06, | |
| "loss": 0.8056, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.869020938873291, | |
| "eval_runtime": 29.9807, | |
| "eval_samples_per_second": 119.477, | |
| "eval_steps_per_second": 5.003, | |
| "step": 34089 | |
| }, | |
| { | |
| "epoch": 33.397870280735724, | |
| "grad_norm": 2.18399977684021, | |
| "learning_rate": 4.951597289448209e-06, | |
| "loss": 0.8193, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 33.88189738625363, | |
| "grad_norm": 0.905960202217102, | |
| "learning_rate": 4.588576960309778e-06, | |
| "loss": 0.8023, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.863983154296875, | |
| "eval_runtime": 29.8457, | |
| "eval_samples_per_second": 120.017, | |
| "eval_steps_per_second": 5.026, | |
| "step": 35122 | |
| }, | |
| { | |
| "epoch": 34.36592449177154, | |
| "grad_norm": 2.841273069381714, | |
| "learning_rate": 4.225556631171346e-06, | |
| "loss": 0.8067, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 34.84995159728945, | |
| "grad_norm": 0.9258007407188416, | |
| "learning_rate": 3.8625363020329145e-06, | |
| "loss": 0.8146, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.870430052280426, | |
| "eval_runtime": 29.8816, | |
| "eval_samples_per_second": 119.873, | |
| "eval_steps_per_second": 5.02, | |
| "step": 36155 | |
| }, | |
| { | |
| "epoch": 35.33397870280736, | |
| "grad_norm": 0.954750120639801, | |
| "learning_rate": 3.499515972894482e-06, | |
| "loss": 0.7988, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 35.81800580832527, | |
| "grad_norm": 1.9225430488586426, | |
| "learning_rate": 3.1364956437560505e-06, | |
| "loss": 0.8079, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8958960175514221, | |
| "eval_runtime": 29.9806, | |
| "eval_samples_per_second": 119.477, | |
| "eval_steps_per_second": 5.003, | |
| "step": 37188 | |
| }, | |
| { | |
| "epoch": 36.30203291384318, | |
| "grad_norm": 1.3240045309066772, | |
| "learning_rate": 2.7734753146176185e-06, | |
| "loss": 0.8054, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 36.78606001936109, | |
| "grad_norm": 5.058154106140137, | |
| "learning_rate": 2.410454985479187e-06, | |
| "loss": 0.8081, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8801546692848206, | |
| "eval_runtime": 29.8881, | |
| "eval_samples_per_second": 119.847, | |
| "eval_steps_per_second": 5.019, | |
| "step": 38221 | |
| }, | |
| { | |
| "epoch": 37.27008712487899, | |
| "grad_norm": 2.5231587886810303, | |
| "learning_rate": 2.047434656340755e-06, | |
| "loss": 0.8066, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 37.7541142303969, | |
| "grad_norm": 1.007765769958496, | |
| "learning_rate": 1.6844143272023235e-06, | |
| "loss": 0.8059, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.890051543712616, | |
| "eval_runtime": 29.9175, | |
| "eval_samples_per_second": 119.729, | |
| "eval_steps_per_second": 5.014, | |
| "step": 39254 | |
| }, | |
| { | |
| "epoch": 38.23814133591481, | |
| "grad_norm": 1.9777470827102661, | |
| "learning_rate": 1.3213939980638917e-06, | |
| "loss": 0.8077, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 38.72216844143272, | |
| "grad_norm": 2.1567304134368896, | |
| "learning_rate": 9.5837366892546e-07, | |
| "loss": 0.8045, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.888158917427063, | |
| "eval_runtime": 29.9571, | |
| "eval_samples_per_second": 119.571, | |
| "eval_steps_per_second": 5.007, | |
| "step": 40287 | |
| }, | |
| { | |
| "epoch": 39.20619554695063, | |
| "grad_norm": 1.4379490613937378, | |
| "learning_rate": 5.953533397870282e-07, | |
| "loss": 0.7972, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 39.690222652468535, | |
| "grad_norm": 1.5997909307479858, | |
| "learning_rate": 2.3233301064859634e-07, | |
| "loss": 0.8024, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_f1": 0.6409826912339475, | |
| "eval_loss": 0.8918312788009644, | |
| "eval_runtime": 29.6532, | |
| "eval_samples_per_second": 120.796, | |
| "eval_steps_per_second": 5.058, | |
| "step": 41320 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "step": 41320, | |
| "total_flos": 2.3084949799922688e+17, | |
| "train_loss": 0.688100525762774, | |
| "train_runtime": 30496.5649, | |
| "train_samples_per_second": 32.49, | |
| "train_steps_per_second": 1.355 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 41320, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.3084949799922688e+17, | |
| "train_batch_size": 24, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |