| { | |
| "best_metric": 0.8473693995166388, | |
| "best_model_checkpoint": "./XLM-V_16-multi-outputs/checkpoint-97000", | |
| "epoch": 17.171180739953975, | |
| "eval_steps": 1000, | |
| "global_step": 97000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17702248185519562, | |
| "grad_norm": 0.6527151465415955, | |
| "learning_rate": 1.770224818551956e-07, | |
| "loss": 0.6947, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.17702248185519562, | |
| "eval_accuracy": 0.5090112516180424, | |
| "eval_f1": 0.6746288353678654, | |
| "eval_loss": 0.6933906078338623, | |
| "eval_precision": 0.5090112516180424, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 48.5534, | |
| "eval_samples_per_second": 206.844, | |
| "eval_steps_per_second": 12.934, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.35404496371039124, | |
| "grad_norm": 0.5621433258056641, | |
| "learning_rate": 3.540449637103912e-07, | |
| "loss": 0.695, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.35404496371039124, | |
| "eval_accuracy": 0.5090112516180424, | |
| "eval_f1": 0.6746288353678654, | |
| "eval_loss": 0.6930156946182251, | |
| "eval_precision": 0.5090112516180424, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 48.572, | |
| "eval_samples_per_second": 206.765, | |
| "eval_steps_per_second": 12.929, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5310674455655868, | |
| "grad_norm": 3.589186906814575, | |
| "learning_rate": 5.310674455655868e-07, | |
| "loss": 0.6932, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5310674455655868, | |
| "eval_accuracy": 0.5623817584387135, | |
| "eval_f1": 0.6990344449770595, | |
| "eval_loss": 0.686276376247406, | |
| "eval_precision": 0.537772626698978, | |
| "eval_recall": 0.9984350547730829, | |
| "eval_runtime": 48.4146, | |
| "eval_samples_per_second": 207.437, | |
| "eval_steps_per_second": 12.971, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.7080899274207825, | |
| "grad_norm": 6.542071342468262, | |
| "learning_rate": 7.080899274207824e-07, | |
| "loss": 0.6861, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7080899274207825, | |
| "eval_accuracy": 0.5879717215971323, | |
| "eval_f1": 0.7084272829763247, | |
| "eval_loss": 0.6729876399040222, | |
| "eval_precision": 0.5536343612334802, | |
| "eval_recall": 0.9833724569640062, | |
| "eval_runtime": 48.5835, | |
| "eval_samples_per_second": 206.716, | |
| "eval_steps_per_second": 12.926, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.885112409275978, | |
| "grad_norm": 7.369508266448975, | |
| "learning_rate": 8.85112409275978e-07, | |
| "loss": 0.6765, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.885112409275978, | |
| "eval_accuracy": 0.6080852334959673, | |
| "eval_f1": 0.7152365793662278, | |
| "eval_loss": 0.6543276309967041, | |
| "eval_precision": 0.5675086107921928, | |
| "eval_recall": 0.9669405320813772, | |
| "eval_runtime": 48.5351, | |
| "eval_samples_per_second": 206.923, | |
| "eval_steps_per_second": 12.939, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.0621348911311737, | |
| "grad_norm": 17.797739028930664, | |
| "learning_rate": 1.0621348911311736e-06, | |
| "loss": 0.6685, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.0621348911311737, | |
| "eval_accuracy": 0.6179428457632181, | |
| "eval_f1": 0.7082794799665476, | |
| "eval_loss": 0.6436656713485718, | |
| "eval_precision": 0.5792811839323467, | |
| "eval_recall": 0.911189358372457, | |
| "eval_runtime": 48.6552, | |
| "eval_samples_per_second": 206.412, | |
| "eval_steps_per_second": 12.907, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.2391573729863692, | |
| "grad_norm": 9.185620307922363, | |
| "learning_rate": 1.2391573729863691e-06, | |
| "loss": 0.6598, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.2391573729863692, | |
| "eval_accuracy": 0.630986756945136, | |
| "eval_f1": 0.7220638968051597, | |
| "eval_loss": 0.6283608675003052, | |
| "eval_precision": 0.5855023108732669, | |
| "eval_recall": 0.9417057902973396, | |
| "eval_runtime": 48.3633, | |
| "eval_samples_per_second": 207.657, | |
| "eval_steps_per_second": 12.985, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.416179854841565, | |
| "grad_norm": 4.287769794464111, | |
| "learning_rate": 1.4161798548415648e-06, | |
| "loss": 0.6587, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.416179854841565, | |
| "eval_accuracy": 0.6417405157821369, | |
| "eval_f1": 0.7242489270386266, | |
| "eval_loss": 0.6228181719779968, | |
| "eval_precision": 0.5953881048387096, | |
| "eval_recall": 0.9242957746478874, | |
| "eval_runtime": 48.577, | |
| "eval_samples_per_second": 206.744, | |
| "eval_steps_per_second": 12.928, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.5932023366967605, | |
| "grad_norm": 6.105784893035889, | |
| "learning_rate": 1.5932023366967605e-06, | |
| "loss": 0.6522, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.5932023366967605, | |
| "eval_accuracy": 0.6528925619834711, | |
| "eval_f1": 0.736228813559322, | |
| "eval_loss": 0.6043413877487183, | |
| "eval_precision": 0.6003208292201382, | |
| "eval_recall": 0.9516823161189358, | |
| "eval_runtime": 48.4503, | |
| "eval_samples_per_second": 207.284, | |
| "eval_steps_per_second": 12.962, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.7702248185519562, | |
| "grad_norm": 2.1733198165893555, | |
| "learning_rate": 1.770224818551956e-06, | |
| "loss": 0.6593, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.7702248185519562, | |
| "eval_accuracy": 0.6620531713631385, | |
| "eval_f1": 0.646531972505728, | |
| "eval_loss": 0.6066622734069824, | |
| "eval_precision": 0.6913140311804009, | |
| "eval_recall": 0.6071987480438185, | |
| "eval_runtime": 48.6257, | |
| "eval_samples_per_second": 206.537, | |
| "eval_steps_per_second": 12.915, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9472473004071515, | |
| "grad_norm": 11.66081714630127, | |
| "learning_rate": 1.9472473004071516e-06, | |
| "loss": 0.6324, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.9472473004071515, | |
| "eval_accuracy": 0.667430050781639, | |
| "eval_f1": 0.6654647435897436, | |
| "eval_loss": 0.5872398018836975, | |
| "eval_precision": 0.6818555008210181, | |
| "eval_recall": 0.6498435054773083, | |
| "eval_runtime": 48.3877, | |
| "eval_samples_per_second": 207.553, | |
| "eval_steps_per_second": 12.979, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.1242697822623473, | |
| "grad_norm": 22.538427352905273, | |
| "learning_rate": 1.9861922464152945e-06, | |
| "loss": 0.6223, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.1242697822623473, | |
| "eval_accuracy": 0.6970028875833915, | |
| "eval_f1": 0.7325305440801617, | |
| "eval_loss": 0.5736498236656189, | |
| "eval_precision": 0.6651237031125299, | |
| "eval_recall": 0.8151408450704225, | |
| "eval_runtime": 48.7219, | |
| "eval_samples_per_second": 206.129, | |
| "eval_steps_per_second": 12.889, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.301292264117543, | |
| "grad_norm": 34.23973846435547, | |
| "learning_rate": 1.9665230817647173e-06, | |
| "loss": 0.6164, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.301292264117543, | |
| "eval_accuracy": 0.6979986059942248, | |
| "eval_f1": 0.7626946248337376, | |
| "eval_loss": 0.562938392162323, | |
| "eval_precision": 0.635545703481549, | |
| "eval_recall": 0.9534428794992176, | |
| "eval_runtime": 48.4533, | |
| "eval_samples_per_second": 207.272, | |
| "eval_steps_per_second": 12.961, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.4783147459727384, | |
| "grad_norm": 12.478341102600098, | |
| "learning_rate": 1.94685391711414e-06, | |
| "loss": 0.5989, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.4783147459727384, | |
| "eval_accuracy": 0.6986956088818083, | |
| "eval_f1": 0.7548209366391184, | |
| "eval_loss": 0.5621292591094971, | |
| "eval_precision": 0.6442600276625173, | |
| "eval_recall": 0.911189358372457, | |
| "eval_runtime": 48.6389, | |
| "eval_samples_per_second": 206.481, | |
| "eval_steps_per_second": 12.911, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.655337227827934, | |
| "grad_norm": 12.369190216064453, | |
| "learning_rate": 1.9271847524635626e-06, | |
| "loss": 0.5746, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.655337227827934, | |
| "eval_accuracy": 0.7115403763815593, | |
| "eval_f1": 0.7359402059976301, | |
| "eval_loss": 0.5231748819351196, | |
| "eval_precision": 0.6890254309609148, | |
| "eval_recall": 0.7897104851330203, | |
| "eval_runtime": 48.3574, | |
| "eval_samples_per_second": 207.683, | |
| "eval_steps_per_second": 12.987, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.83235970968313, | |
| "grad_norm": 20.652965545654297, | |
| "learning_rate": 1.9075155878129855e-06, | |
| "loss": 0.5257, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.83235970968313, | |
| "eval_accuracy": 0.7224932789007269, | |
| "eval_f1": 0.7712386111795124, | |
| "eval_loss": 0.4791330397129059, | |
| "eval_precision": 0.6644039032668646, | |
| "eval_recall": 0.9190140845070423, | |
| "eval_runtime": 48.5117, | |
| "eval_samples_per_second": 207.022, | |
| "eval_steps_per_second": 12.945, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.009382191538325, | |
| "grad_norm": 12.337203025817871, | |
| "learning_rate": 1.8878464231624083e-06, | |
| "loss": 0.4955, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.009382191538325, | |
| "eval_accuracy": 0.7429055063228119, | |
| "eval_f1": 0.7466143277723258, | |
| "eval_loss": 0.4369197189807892, | |
| "eval_precision": 0.7491138243402915, | |
| "eval_recall": 0.744131455399061, | |
| "eval_runtime": 48.4884, | |
| "eval_samples_per_second": 207.122, | |
| "eval_steps_per_second": 12.952, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.186404673393521, | |
| "grad_norm": 78.35604858398438, | |
| "learning_rate": 1.868177258511831e-06, | |
| "loss": 0.4701, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.186404673393521, | |
| "eval_accuracy": 0.7424076471173952, | |
| "eval_f1": 0.7794731906913307, | |
| "eval_loss": 0.4281806945800781, | |
| "eval_precision": 0.6907387822933978, | |
| "eval_recall": 0.8943661971830986, | |
| "eval_runtime": 48.4785, | |
| "eval_samples_per_second": 207.164, | |
| "eval_steps_per_second": 12.954, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.3634271552487167, | |
| "grad_norm": 12.168407440185547, | |
| "learning_rate": 1.8485080938612536e-06, | |
| "loss": 0.4503, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.3634271552487167, | |
| "eval_accuracy": 0.7501742507218958, | |
| "eval_f1": 0.73006993006993, | |
| "eval_loss": 0.4149376451969147, | |
| "eval_precision": 0.8111403299067654, | |
| "eval_recall": 0.6637323943661971, | |
| "eval_runtime": 48.505, | |
| "eval_samples_per_second": 207.051, | |
| "eval_steps_per_second": 12.947, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.5404496371039125, | |
| "grad_norm": 22.702987670898438, | |
| "learning_rate": 1.8288389292106762e-06, | |
| "loss": 0.4408, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.5404496371039125, | |
| "eval_accuracy": 0.7519665438613959, | |
| "eval_f1": 0.7831084022638224, | |
| "eval_loss": 0.4072709083557129, | |
| "eval_precision": 0.7056331398085673, | |
| "eval_recall": 0.8796948356807511, | |
| "eval_runtime": 48.5087, | |
| "eval_samples_per_second": 207.035, | |
| "eval_steps_per_second": 12.946, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.717472118959108, | |
| "grad_norm": 11.862476348876953, | |
| "learning_rate": 1.809169764560099e-06, | |
| "loss": 0.4318, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.717472118959108, | |
| "eval_accuracy": 0.7565468485512297, | |
| "eval_f1": 0.7653325655053268, | |
| "eval_loss": 0.398566335439682, | |
| "eval_precision": 0.751271905031091, | |
| "eval_recall": 0.7799295774647887, | |
| "eval_runtime": 48.5139, | |
| "eval_samples_per_second": 207.013, | |
| "eval_steps_per_second": 12.945, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.8944946008143035, | |
| "grad_norm": 5.476343631744385, | |
| "learning_rate": 1.7895005999095219e-06, | |
| "loss": 0.4275, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.8944946008143035, | |
| "eval_accuracy": 0.7611271532410634, | |
| "eval_f1": 0.7729294841457643, | |
| "eval_loss": 0.3958267569541931, | |
| "eval_precision": 0.7487621492756281, | |
| "eval_recall": 0.7987089201877934, | |
| "eval_runtime": 48.4857, | |
| "eval_samples_per_second": 207.133, | |
| "eval_steps_per_second": 12.952, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 4.071517082669499, | |
| "grad_norm": 4.330970764160156, | |
| "learning_rate": 1.7698314352589445e-06, | |
| "loss": 0.4162, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 4.071517082669499, | |
| "eval_accuracy": 0.7533605496365627, | |
| "eval_f1": 0.7667827888146126, | |
| "eval_loss": 0.3944496214389801, | |
| "eval_precision": 0.7391541114539844, | |
| "eval_recall": 0.7965571205007824, | |
| "eval_runtime": 48.5603, | |
| "eval_samples_per_second": 206.815, | |
| "eval_steps_per_second": 12.932, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 4.248539564524695, | |
| "grad_norm": 4.463377475738525, | |
| "learning_rate": 1.7501622706083671e-06, | |
| "loss": 0.4158, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 4.248539564524695, | |
| "eval_accuracy": 0.7555511301403963, | |
| "eval_f1": 0.7441908929873919, | |
| "eval_loss": 0.3938951790332794, | |
| "eval_precision": 0.7962095875139353, | |
| "eval_recall": 0.6985524256651017, | |
| "eval_runtime": 48.4747, | |
| "eval_samples_per_second": 207.18, | |
| "eval_steps_per_second": 12.955, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 4.42556204637989, | |
| "grad_norm": 102.01718139648438, | |
| "learning_rate": 1.7304931059577898e-06, | |
| "loss": 0.412, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 4.42556204637989, | |
| "eval_accuracy": 0.7532609777954794, | |
| "eval_f1": 0.79493545183714, | |
| "eval_loss": 0.404395192861557, | |
| "eval_precision": 0.6888984509466437, | |
| "eval_recall": 0.9395539906103286, | |
| "eval_runtime": 48.4966, | |
| "eval_samples_per_second": 207.087, | |
| "eval_steps_per_second": 12.949, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 4.602584528235086, | |
| "grad_norm": 4.4260969161987305, | |
| "learning_rate": 1.7108239413072126e-06, | |
| "loss": 0.406, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 4.602584528235086, | |
| "eval_accuracy": 0.7578412824853131, | |
| "eval_f1": 0.7183229094278434, | |
| "eval_loss": 0.3970828354358673, | |
| "eval_precision": 0.880465644520159, | |
| "eval_recall": 0.6066118935837246, | |
| "eval_runtime": 48.4882, | |
| "eval_samples_per_second": 207.122, | |
| "eval_steps_per_second": 12.952, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 4.779607010090281, | |
| "grad_norm": 9.056214332580566, | |
| "learning_rate": 1.6911547766566355e-06, | |
| "loss": 0.4044, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 4.779607010090281, | |
| "eval_accuracy": 0.7609280095588967, | |
| "eval_f1": 0.7647231749142577, | |
| "eval_loss": 0.3946742117404938, | |
| "eval_precision": 0.7661496171215394, | |
| "eval_recall": 0.763302034428795, | |
| "eval_runtime": 48.479, | |
| "eval_samples_per_second": 207.162, | |
| "eval_steps_per_second": 12.954, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 4.956629491945477, | |
| "grad_norm": 70.51245880126953, | |
| "learning_rate": 1.671485612006058e-06, | |
| "loss": 0.407, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 4.956629491945477, | |
| "eval_accuracy": 0.7619237279697302, | |
| "eval_f1": 0.7924659317767555, | |
| "eval_loss": 0.3901920020580292, | |
| "eval_precision": 0.7122796068029333, | |
| "eval_recall": 0.8929968701095462, | |
| "eval_runtime": 48.4988, | |
| "eval_samples_per_second": 207.077, | |
| "eval_steps_per_second": 12.949, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 5.1336519738006725, | |
| "grad_norm": 1.7648775577545166, | |
| "learning_rate": 1.6518164473554807e-06, | |
| "loss": 0.3916, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 5.1336519738006725, | |
| "eval_accuracy": 0.7610275813999801, | |
| "eval_f1": 0.7391304347826086, | |
| "eval_loss": 0.38304734230041504, | |
| "eval_precision": 0.8317025440313112, | |
| "eval_recall": 0.6651017214397497, | |
| "eval_runtime": 48.5026, | |
| "eval_samples_per_second": 207.061, | |
| "eval_steps_per_second": 12.948, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 5.310674455655868, | |
| "grad_norm": 5.286086559295654, | |
| "learning_rate": 1.6321472827049034e-06, | |
| "loss": 0.4001, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 5.310674455655868, | |
| "eval_accuracy": 0.764512595837897, | |
| "eval_f1": 0.7388183324130315, | |
| "eval_loss": 0.3838052451610565, | |
| "eval_precision": 0.8483388283033223, | |
| "eval_recall": 0.6543427230046949, | |
| "eval_runtime": 48.8022, | |
| "eval_samples_per_second": 205.79, | |
| "eval_steps_per_second": 12.868, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 5.487696937511064, | |
| "grad_norm": 22.618858337402344, | |
| "learning_rate": 1.612478118054326e-06, | |
| "loss": 0.3886, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 5.487696937511064, | |
| "eval_accuracy": 0.7602310066713134, | |
| "eval_f1": 0.8046088932164882, | |
| "eval_loss": 0.3872602581977844, | |
| "eval_precision": 0.687465335551858, | |
| "eval_recall": 0.9698748043818466, | |
| "eval_runtime": 48.4944, | |
| "eval_samples_per_second": 207.096, | |
| "eval_steps_per_second": 12.95, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 5.66471941936626, | |
| "grad_norm": 11.363678932189941, | |
| "learning_rate": 1.5928089534037488e-06, | |
| "loss": 0.3802, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 5.66471941936626, | |
| "eval_accuracy": 0.7706860499850642, | |
| "eval_f1": 0.800138852729324, | |
| "eval_loss": 0.37955331802368164, | |
| "eval_precision": 0.7190765871158945, | |
| "eval_recall": 0.9017996870109546, | |
| "eval_runtime": 48.4665, | |
| "eval_samples_per_second": 207.215, | |
| "eval_steps_per_second": 12.957, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 5.8417419012214555, | |
| "grad_norm": 18.808916091918945, | |
| "learning_rate": 1.5731397887531717e-06, | |
| "loss": 0.383, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 5.8417419012214555, | |
| "eval_accuracy": 0.7748680673105646, | |
| "eval_f1": 0.8042932571626418, | |
| "eval_loss": 0.3785202205181122, | |
| "eval_precision": 0.7213165657506598, | |
| "eval_recall": 0.9088419405320813, | |
| "eval_runtime": 48.5221, | |
| "eval_samples_per_second": 206.978, | |
| "eval_steps_per_second": 12.943, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 6.01876438307665, | |
| "grad_norm": 3.9943904876708984, | |
| "learning_rate": 1.5534706241025943e-06, | |
| "loss": 0.3833, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 6.01876438307665, | |
| "eval_accuracy": 0.7772577914965648, | |
| "eval_f1": 0.787296757630503, | |
| "eval_loss": 0.3748994767665863, | |
| "eval_precision": 0.7659574468085106, | |
| "eval_recall": 0.8098591549295775, | |
| "eval_runtime": 48.5676, | |
| "eval_samples_per_second": 206.784, | |
| "eval_steps_per_second": 12.93, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 6.195786864931846, | |
| "grad_norm": 5.812683582305908, | |
| "learning_rate": 1.533801459452017e-06, | |
| "loss": 0.3695, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 6.195786864931846, | |
| "eval_accuracy": 0.776759932291148, | |
| "eval_f1": 0.8093537414965987, | |
| "eval_loss": 0.38256847858428955, | |
| "eval_precision": 0.7158543922984356, | |
| "eval_recall": 0.9309467918622848, | |
| "eval_runtime": 48.4235, | |
| "eval_samples_per_second": 207.399, | |
| "eval_steps_per_second": 12.969, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 6.372809346787042, | |
| "grad_norm": 3.7167677879333496, | |
| "learning_rate": 1.5141322948014396e-06, | |
| "loss": 0.3732, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 6.372809346787042, | |
| "eval_accuracy": 0.7745693517873146, | |
| "eval_f1": 0.8088806348134391, | |
| "eval_loss": 0.3752972483634949, | |
| "eval_precision": 0.7114642114642115, | |
| "eval_recall": 0.937206572769953, | |
| "eval_runtime": 48.483, | |
| "eval_samples_per_second": 207.145, | |
| "eval_steps_per_second": 12.953, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 6.549831828642238, | |
| "grad_norm": 16.73068618774414, | |
| "learning_rate": 1.4944631301508624e-06, | |
| "loss": 0.3675, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 6.549831828642238, | |
| "eval_accuracy": 0.7851239669421488, | |
| "eval_f1": 0.7800652262535671, | |
| "eval_loss": 0.3612446188926697, | |
| "eval_precision": 0.8142553191489361, | |
| "eval_recall": 0.7486306729264476, | |
| "eval_runtime": 48.4403, | |
| "eval_samples_per_second": 207.327, | |
| "eval_steps_per_second": 12.964, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 6.726854310497433, | |
| "grad_norm": 11.350728988647461, | |
| "learning_rate": 1.4747939655002853e-06, | |
| "loss": 0.3631, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 6.726854310497433, | |
| "eval_accuracy": 0.7871154037638156, | |
| "eval_f1": 0.7887769215570045, | |
| "eval_loss": 0.3669988214969635, | |
| "eval_precision": 0.7968063872255489, | |
| "eval_recall": 0.7809076682316118, | |
| "eval_runtime": 48.6242, | |
| "eval_samples_per_second": 206.543, | |
| "eval_steps_per_second": 12.915, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 6.903876792352628, | |
| "grad_norm": 19.71718406677246, | |
| "learning_rate": 1.455124800849708e-06, | |
| "loss": 0.3619, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 6.903876792352628, | |
| "eval_accuracy": 0.7860201135118988, | |
| "eval_f1": 0.8045475216007276, | |
| "eval_loss": 0.3651063144207001, | |
| "eval_precision": 0.7518272989971103, | |
| "eval_recall": 0.8652190923317684, | |
| "eval_runtime": 48.6286, | |
| "eval_samples_per_second": 206.524, | |
| "eval_steps_per_second": 12.914, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 7.080899274207824, | |
| "grad_norm": 25.673242568969727, | |
| "learning_rate": 1.4354556361991305e-06, | |
| "loss": 0.3587, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 7.080899274207824, | |
| "eval_accuracy": 0.795977297620233, | |
| "eval_f1": 0.7845652402481338, | |
| "eval_loss": 0.35970279574394226, | |
| "eval_precision": 0.8481473062059559, | |
| "eval_recall": 0.7298513302034428, | |
| "eval_runtime": 48.507, | |
| "eval_samples_per_second": 207.042, | |
| "eval_steps_per_second": 12.947, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 7.25792175606302, | |
| "grad_norm": 19.93338966369629, | |
| "learning_rate": 1.4157864715485532e-06, | |
| "loss": 0.349, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 7.25792175606302, | |
| "eval_accuracy": 0.7883102658568157, | |
| "eval_f1": 0.7995096190116937, | |
| "eval_loss": 0.3596351444721222, | |
| "eval_precision": 0.7718499635833941, | |
| "eval_recall": 0.829225352112676, | |
| "eval_runtime": 48.3456, | |
| "eval_samples_per_second": 207.734, | |
| "eval_steps_per_second": 12.99, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 7.434944237918216, | |
| "grad_norm": 7.293708801269531, | |
| "learning_rate": 1.396117306897976e-06, | |
| "loss": 0.3402, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 7.434944237918216, | |
| "eval_accuracy": 0.7938862889574828, | |
| "eval_f1": 0.7982849347105827, | |
| "eval_loss": 0.3619771897792816, | |
| "eval_precision": 0.7953398058252428, | |
| "eval_recall": 0.8012519561815337, | |
| "eval_runtime": 48.7896, | |
| "eval_samples_per_second": 205.843, | |
| "eval_steps_per_second": 12.872, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 7.611966719773411, | |
| "grad_norm": 82.438720703125, | |
| "learning_rate": 1.3764481422473986e-06, | |
| "loss": 0.3448, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 7.611966719773411, | |
| "eval_accuracy": 0.7969730160310664, | |
| "eval_f1": 0.8135345221764975, | |
| "eval_loss": 0.3580198884010315, | |
| "eval_precision": 0.7638674222909153, | |
| "eval_recall": 0.8701095461658842, | |
| "eval_runtime": 48.395, | |
| "eval_samples_per_second": 207.522, | |
| "eval_steps_per_second": 12.977, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 7.788989201628607, | |
| "grad_norm": 13.257980346679688, | |
| "learning_rate": 1.3567789775968215e-06, | |
| "loss": 0.336, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 7.788989201628607, | |
| "eval_accuracy": 0.7987653091705665, | |
| "eval_f1": 0.7967005331455588, | |
| "eval_loss": 0.3453768789768219, | |
| "eval_precision": 0.8200455580865603, | |
| "eval_recall": 0.7746478873239436, | |
| "eval_runtime": 48.664, | |
| "eval_samples_per_second": 206.374, | |
| "eval_steps_per_second": 12.905, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 7.966011683483803, | |
| "grad_norm": 11.088878631591797, | |
| "learning_rate": 1.3371098129462441e-06, | |
| "loss": 0.336, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 7.966011683483803, | |
| "eval_accuracy": 0.8018520362441501, | |
| "eval_f1": 0.8217484772482981, | |
| "eval_loss": 0.3519151210784912, | |
| "eval_precision": 0.7579312623925974, | |
| "eval_recall": 0.8973004694835681, | |
| "eval_runtime": 48.4613, | |
| "eval_samples_per_second": 207.237, | |
| "eval_steps_per_second": 12.959, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 8.143034165338998, | |
| "grad_norm": 10.603625297546387, | |
| "learning_rate": 1.3174406482956667e-06, | |
| "loss": 0.3307, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 8.143034165338998, | |
| "eval_accuracy": 0.8025490391317336, | |
| "eval_f1": 0.8225185715564307, | |
| "eval_loss": 0.3641522526741028, | |
| "eval_precision": 0.7581257218280811, | |
| "eval_recall": 0.8988654147104851, | |
| "eval_runtime": 48.5668, | |
| "eval_samples_per_second": 206.788, | |
| "eval_steps_per_second": 12.931, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 8.320056647194194, | |
| "grad_norm": 45.665870666503906, | |
| "learning_rate": 1.2977714836450896e-06, | |
| "loss": 0.3192, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 8.320056647194194, | |
| "eval_accuracy": 0.8047396196355671, | |
| "eval_f1": 0.8228705627314605, | |
| "eval_loss": 0.3606601357460022, | |
| "eval_precision": 0.764389998321866, | |
| "eval_recall": 0.8910406885758998, | |
| "eval_runtime": 48.3992, | |
| "eval_samples_per_second": 207.504, | |
| "eval_steps_per_second": 12.975, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 8.49707912904939, | |
| "grad_norm": 45.48286819458008, | |
| "learning_rate": 1.2781023189945122e-06, | |
| "loss": 0.325, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 8.49707912904939, | |
| "eval_accuracy": 0.8093199243254008, | |
| "eval_f1": 0.802963267825908, | |
| "eval_loss": 0.34560200572013855, | |
| "eval_precision": 0.846971999131756, | |
| "eval_recall": 0.763302034428795, | |
| "eval_runtime": 48.6156, | |
| "eval_samples_per_second": 206.58, | |
| "eval_steps_per_second": 12.918, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 8.674101610904584, | |
| "grad_norm": 12.470772743225098, | |
| "learning_rate": 1.258433154343935e-06, | |
| "loss": 0.32, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 8.674101610904584, | |
| "eval_accuracy": 0.8102160708951508, | |
| "eval_f1": 0.8108751736455646, | |
| "eval_loss": 0.3369452655315399, | |
| "eval_precision": 0.8227950060410794, | |
| "eval_recall": 0.7992957746478874, | |
| "eval_runtime": 48.3828, | |
| "eval_samples_per_second": 207.574, | |
| "eval_steps_per_second": 12.98, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 8.85112409275978, | |
| "grad_norm": 9.075933456420898, | |
| "learning_rate": 1.2387639896933577e-06, | |
| "loss": 0.316, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 8.85112409275978, | |
| "eval_accuracy": 0.8132032261276511, | |
| "eval_f1": 0.8101214574898785, | |
| "eval_loss": 0.34877920150756836, | |
| "eval_precision": 0.8393456375838926, | |
| "eval_recall": 0.7828638497652582, | |
| "eval_runtime": 48.5199, | |
| "eval_samples_per_second": 206.987, | |
| "eval_steps_per_second": 12.943, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 9.028146574614976, | |
| "grad_norm": 19.899991989135742, | |
| "learning_rate": 1.2190948250427803e-06, | |
| "loss": 0.3202, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 9.028146574614976, | |
| "eval_accuracy": 0.8114109329881509, | |
| "eval_f1": 0.8094567404426559, | |
| "eval_loss": 0.339097797870636, | |
| "eval_precision": 0.8332642916321458, | |
| "eval_recall": 0.7869718309859155, | |
| "eval_runtime": 48.4752, | |
| "eval_samples_per_second": 207.178, | |
| "eval_steps_per_second": 12.955, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 9.205169056470172, | |
| "grad_norm": 12.141376495361328, | |
| "learning_rate": 1.199425660392203e-06, | |
| "loss": 0.3081, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 9.205169056470172, | |
| "eval_accuracy": 0.8111122174649009, | |
| "eval_f1": 0.8052561338671594, | |
| "eval_loss": 0.32928845286369324, | |
| "eval_precision": 0.8472672283430547, | |
| "eval_recall": 0.7672143974960877, | |
| "eval_runtime": 48.4715, | |
| "eval_samples_per_second": 207.194, | |
| "eval_steps_per_second": 12.956, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 9.382191538325367, | |
| "grad_norm": 10.343143463134766, | |
| "learning_rate": 1.1797564957416258e-06, | |
| "loss": 0.3036, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 9.382191538325367, | |
| "eval_accuracy": 0.8130040824454844, | |
| "eval_f1": 0.8091463414634147, | |
| "eval_loss": 0.3455412685871124, | |
| "eval_precision": 0.8420050761421319, | |
| "eval_recall": 0.778755868544601, | |
| "eval_runtime": 48.4453, | |
| "eval_samples_per_second": 207.306, | |
| "eval_steps_per_second": 12.963, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 9.559214020180562, | |
| "grad_norm": 15.281163215637207, | |
| "learning_rate": 1.1600873310910486e-06, | |
| "loss": 0.3025, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 9.559214020180562, | |
| "eval_accuracy": 0.8187792492283182, | |
| "eval_f1": 0.828883038736367, | |
| "eval_loss": 0.3434953987598419, | |
| "eval_precision": 0.7979724837074583, | |
| "eval_recall": 0.8622848200312989, | |
| "eval_runtime": 48.4882, | |
| "eval_samples_per_second": 207.123, | |
| "eval_steps_per_second": 12.952, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 9.736236502035759, | |
| "grad_norm": 10.584362983703613, | |
| "learning_rate": 1.1404181664404713e-06, | |
| "loss": 0.2983, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 9.736236502035759, | |
| "eval_accuracy": 0.8190779647515682, | |
| "eval_f1": 0.8283419933868682, | |
| "eval_loss": 0.3416574001312256, | |
| "eval_precision": 0.8010232048236798, | |
| "eval_recall": 0.8575899843505478, | |
| "eval_runtime": 48.4574, | |
| "eval_samples_per_second": 207.254, | |
| "eval_steps_per_second": 12.96, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 9.913258983890954, | |
| "grad_norm": 18.094242095947266, | |
| "learning_rate": 1.120749001789894e-06, | |
| "loss": 0.2959, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 9.913258983890954, | |
| "eval_accuracy": 0.8174848152942348, | |
| "eval_f1": 0.8205228630177225, | |
| "eval_loss": 0.3372325301170349, | |
| "eval_precision": 0.8214075671436973, | |
| "eval_recall": 0.8196400625978091, | |
| "eval_runtime": 48.5309, | |
| "eval_samples_per_second": 206.94, | |
| "eval_steps_per_second": 12.94, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 10.09028146574615, | |
| "grad_norm": 7.2225422859191895, | |
| "learning_rate": 1.1010798371393165e-06, | |
| "loss": 0.2917, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 10.09028146574615, | |
| "eval_accuracy": 0.8219655481429852, | |
| "eval_f1": 0.8156320890905341, | |
| "eval_loss": 0.3421269953250885, | |
| "eval_precision": 0.8624073266463149, | |
| "eval_recall": 0.7736697965571205, | |
| "eval_runtime": 48.4733, | |
| "eval_samples_per_second": 207.186, | |
| "eval_steps_per_second": 12.956, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 10.267303947601345, | |
| "grad_norm": 21.429712295532227, | |
| "learning_rate": 1.0814106724887394e-06, | |
| "loss": 0.2856, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 10.267303947601345, | |
| "eval_accuracy": 0.8219655481429852, | |
| "eval_f1": 0.8340141106572596, | |
| "eval_loss": 0.3443463146686554, | |
| "eval_precision": 0.7936395759717314, | |
| "eval_recall": 0.878716744913928, | |
| "eval_runtime": 48.4616, | |
| "eval_samples_per_second": 207.236, | |
| "eval_steps_per_second": 12.959, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 10.444326429456542, | |
| "grad_norm": 89.16828918457031, | |
| "learning_rate": 1.061741507838162e-06, | |
| "loss": 0.2855, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 10.444326429456542, | |
| "eval_accuracy": 0.8195758239569849, | |
| "eval_f1": 0.8235981308411215, | |
| "eval_loss": 0.3458768427371979, | |
| "eval_precision": 0.8197674418604651, | |
| "eval_recall": 0.8274647887323944, | |
| "eval_runtime": 48.4454, | |
| "eval_samples_per_second": 207.305, | |
| "eval_steps_per_second": 12.963, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 10.621348911311737, | |
| "grad_norm": 6.35772705078125, | |
| "learning_rate": 1.0420723431875849e-06, | |
| "loss": 0.2869, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 10.621348911311737, | |
| "eval_accuracy": 0.8268445683560689, | |
| "eval_f1": 0.8420957050758194, | |
| "eval_loss": 0.3441038131713867, | |
| "eval_precision": 0.7857990171157431, | |
| "eval_recall": 0.9070813771517997, | |
| "eval_runtime": 48.5327, | |
| "eval_samples_per_second": 206.932, | |
| "eval_steps_per_second": 12.94, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 10.798371393166931, | |
| "grad_norm": 13.831587791442871, | |
| "learning_rate": 1.0224031785370075e-06, | |
| "loss": 0.2902, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 10.798371393166931, | |
| "eval_accuracy": 0.8228616947127353, | |
| "eval_f1": 0.8143200083498591, | |
| "eval_loss": 0.3371775150299072, | |
| "eval_precision": 0.8729022152606847, | |
| "eval_recall": 0.7631064162754304, | |
| "eval_runtime": 48.4429, | |
| "eval_samples_per_second": 207.316, | |
| "eval_steps_per_second": 12.964, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 10.975393875022128, | |
| "grad_norm": 47.082096099853516, | |
| "learning_rate": 1.0027340138864301e-06, | |
| "loss": 0.2886, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 10.975393875022128, | |
| "eval_accuracy": 0.8266454246739022, | |
| "eval_f1": 0.8348980559506876, | |
| "eval_loss": 0.32535311579704285, | |
| "eval_precision": 0.8102337566721884, | |
| "eval_recall": 0.8611111111111112, | |
| "eval_runtime": 48.5202, | |
| "eval_samples_per_second": 206.986, | |
| "eval_steps_per_second": 12.943, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 11.152416356877323, | |
| "grad_norm": 14.556977272033691, | |
| "learning_rate": 9.83064849235853e-07, | |
| "loss": 0.2788, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 11.152416356877323, | |
| "eval_accuracy": 0.8296325799064025, | |
| "eval_f1": 0.8395987625386707, | |
| "eval_loss": 0.3404853045940399, | |
| "eval_precision": 0.8061206120612061, | |
| "eval_recall": 0.8759780907668232, | |
| "eval_runtime": 48.4849, | |
| "eval_samples_per_second": 207.137, | |
| "eval_steps_per_second": 12.952, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 11.32943883873252, | |
| "grad_norm": 49.14487075805664, | |
| "learning_rate": 9.633956845852756e-07, | |
| "loss": 0.2722, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 11.32943883873252, | |
| "eval_accuracy": 0.8275415712436522, | |
| "eval_f1": 0.8202573682025737, | |
| "eval_loss": 0.3554268181324005, | |
| "eval_precision": 0.8735632183908046, | |
| "eval_recall": 0.7730829420970265, | |
| "eval_runtime": 48.5073, | |
| "eval_samples_per_second": 207.041, | |
| "eval_steps_per_second": 12.947, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 11.506461320587714, | |
| "grad_norm": 75.99839782714844, | |
| "learning_rate": 9.437265199346983e-07, | |
| "loss": 0.2727, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 11.506461320587714, | |
| "eval_accuracy": 0.8255501344219854, | |
| "eval_f1": 0.8319263238679969, | |
| "eval_loss": 0.3575954735279083, | |
| "eval_precision": 0.8162650602409639, | |
| "eval_recall": 0.8482003129890454, | |
| "eval_runtime": 48.4953, | |
| "eval_samples_per_second": 207.092, | |
| "eval_steps_per_second": 12.95, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 11.683483802442911, | |
| "grad_norm": 10.247947692871094, | |
| "learning_rate": 9.24057355284121e-07, | |
| "loss": 0.2797, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 11.683483802442911, | |
| "eval_accuracy": 0.8265458528328189, | |
| "eval_f1": 0.8270796108794918, | |
| "eval_loss": 0.32501116394996643, | |
| "eval_precision": 0.8395808141878275, | |
| "eval_recall": 0.814945226917058, | |
| "eval_runtime": 48.5218, | |
| "eval_samples_per_second": 206.979, | |
| "eval_steps_per_second": 12.943, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 11.860506284298106, | |
| "grad_norm": 14.356186866760254, | |
| "learning_rate": 9.043881906335438e-07, | |
| "loss": 0.2775, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 11.860506284298106, | |
| "eval_accuracy": 0.8279398586079857, | |
| "eval_f1": 0.8267148014440433, | |
| "eval_loss": 0.34262222051620483, | |
| "eval_precision": 0.8481481481481481, | |
| "eval_recall": 0.8063380281690141, | |
| "eval_runtime": 48.4239, | |
| "eval_samples_per_second": 207.397, | |
| "eval_steps_per_second": 12.969, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 12.0375287661533, | |
| "grad_norm": 55.23761749267578, | |
| "learning_rate": 8.847190259829664e-07, | |
| "loss": 0.2753, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 12.0375287661533, | |
| "eval_accuracy": 0.8288360051777357, | |
| "eval_f1": 0.8412302576891105, | |
| "eval_loss": 0.3272722363471985, | |
| "eval_precision": 0.7968503937007874, | |
| "eval_recall": 0.8908450704225352, | |
| "eval_runtime": 48.5309, | |
| "eval_samples_per_second": 206.94, | |
| "eval_steps_per_second": 12.94, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 12.214551248008497, | |
| "grad_norm": 8.974770545959473, | |
| "learning_rate": 8.650498613323892e-07, | |
| "loss": 0.2625, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 12.214551248008497, | |
| "eval_accuracy": 0.8281390022901524, | |
| "eval_f1": 0.8395910780669145, | |
| "eval_loss": 0.33536896109580994, | |
| "eval_precision": 0.7997521246458924, | |
| "eval_recall": 0.8836071987480438, | |
| "eval_runtime": 48.4526, | |
| "eval_samples_per_second": 207.275, | |
| "eval_steps_per_second": 12.961, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 12.391573729863692, | |
| "grad_norm": 6.9588446617126465, | |
| "learning_rate": 8.453806966818119e-07, | |
| "loss": 0.2593, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 12.391573729863692, | |
| "eval_accuracy": 0.8278402867669024, | |
| "eval_f1": 0.8303071940327804, | |
| "eval_loss": 0.33842170238494873, | |
| "eval_precision": 0.8331691944061453, | |
| "eval_recall": 0.8274647887323944, | |
| "eval_runtime": 48.5216, | |
| "eval_samples_per_second": 206.98, | |
| "eval_steps_per_second": 12.943, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 12.568596211718889, | |
| "grad_norm": 2.9586853981018066, | |
| "learning_rate": 8.257115320312346e-07, | |
| "loss": 0.2716, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 12.568596211718889, | |
| "eval_accuracy": 0.8300308672707358, | |
| "eval_f1": 0.8259051504334524, | |
| "eval_loss": 0.3227674663066864, | |
| "eval_precision": 0.8627743447688047, | |
| "eval_recall": 0.792057902973396, | |
| "eval_runtime": 48.4717, | |
| "eval_samples_per_second": 207.193, | |
| "eval_steps_per_second": 12.956, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 12.745618693574084, | |
| "grad_norm": 41.94707489013672, | |
| "learning_rate": 8.060423673806573e-07, | |
| "loss": 0.2673, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 12.745618693574084, | |
| "eval_accuracy": 0.8310265856815693, | |
| "eval_f1": 0.8221733207586712, | |
| "eval_loss": 0.3358391225337982, | |
| "eval_precision": 0.8853531934100655, | |
| "eval_recall": 0.7674100156494522, | |
| "eval_runtime": 48.4704, | |
| "eval_samples_per_second": 207.199, | |
| "eval_steps_per_second": 12.956, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 12.922641175429279, | |
| "grad_norm": 29.572885513305664, | |
| "learning_rate": 7.8637320273008e-07, | |
| "loss": 0.2656, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 12.922641175429279, | |
| "eval_accuracy": 0.8303295827939858, | |
| "eval_f1": 0.8300757877941763, | |
| "eval_loss": 0.33422765135765076, | |
| "eval_precision": 0.8466232709519935, | |
| "eval_recall": 0.8141627543035994, | |
| "eval_runtime": 48.4871, | |
| "eval_samples_per_second": 207.127, | |
| "eval_steps_per_second": 12.952, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 13.099663657284475, | |
| "grad_norm": 23.2849178314209, | |
| "learning_rate": 7.667040380795027e-07, | |
| "loss": 0.2536, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 13.099663657284475, | |
| "eval_accuracy": 0.8306282983172358, | |
| "eval_f1": 0.8436149673623242, | |
| "eval_loss": 0.3436105251312256, | |
| "eval_precision": 0.795836947094536, | |
| "eval_recall": 0.8974960876369327, | |
| "eval_runtime": 48.5424, | |
| "eval_samples_per_second": 206.891, | |
| "eval_steps_per_second": 12.937, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 13.27668613913967, | |
| "grad_norm": 8.842978477478027, | |
| "learning_rate": 7.470348734289255e-07, | |
| "loss": 0.2591, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 13.27668613913967, | |
| "eval_accuracy": 0.823956984964652, | |
| "eval_f1": 0.8267006469319741, | |
| "eval_loss": 0.34493395686149597, | |
| "eval_precision": 0.8284872298624755, | |
| "eval_recall": 0.8249217527386542, | |
| "eval_runtime": 48.4584, | |
| "eval_samples_per_second": 207.25, | |
| "eval_steps_per_second": 12.96, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 13.453708620994867, | |
| "grad_norm": 52.21194839477539, | |
| "learning_rate": 7.273657087783481e-07, | |
| "loss": 0.2558, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 13.453708620994867, | |
| "eval_accuracy": 0.8287364333366524, | |
| "eval_f1": 0.833106928003105, | |
| "eval_loss": 0.35551726818084717, | |
| "eval_precision": 0.826530612244898, | |
| "eval_recall": 0.8397887323943662, | |
| "eval_runtime": 48.5138, | |
| "eval_samples_per_second": 207.013, | |
| "eval_steps_per_second": 12.945, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 13.630731102850062, | |
| "grad_norm": 8.401360511779785, | |
| "learning_rate": 7.076965441277709e-07, | |
| "loss": 0.2534, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 13.630731102850062, | |
| "eval_accuracy": 0.8238574131235686, | |
| "eval_f1": 0.8269249584189414, | |
| "eval_loss": 0.3419524133205414, | |
| "eval_precision": 0.8271677431982776, | |
| "eval_recall": 0.8266823161189358, | |
| "eval_runtime": 48.4633, | |
| "eval_samples_per_second": 207.229, | |
| "eval_steps_per_second": 12.958, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 13.807753584705258, | |
| "grad_norm": 19.153236389160156, | |
| "learning_rate": 6.880273794771936e-07, | |
| "loss": 0.2574, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 13.807753584705258, | |
| "eval_accuracy": 0.8307278701583193, | |
| "eval_f1": 0.8393194706994329, | |
| "eval_loss": 0.3600776791572571, | |
| "eval_precision": 0.8119970738844184, | |
| "eval_recall": 0.8685446009389671, | |
| "eval_runtime": 48.5086, | |
| "eval_samples_per_second": 207.035, | |
| "eval_steps_per_second": 12.946, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 13.984776066560453, | |
| "grad_norm": 5.686819553375244, | |
| "learning_rate": 6.683582148266162e-07, | |
| "loss": 0.2605, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 13.984776066560453, | |
| "eval_accuracy": 0.8274419994025689, | |
| "eval_f1": 0.8354382299876555, | |
| "eval_loss": 0.34466180205345154, | |
| "eval_precision": 0.8117733899243403, | |
| "eval_recall": 0.8605242566510172, | |
| "eval_runtime": 48.4951, | |
| "eval_samples_per_second": 207.093, | |
| "eval_steps_per_second": 12.95, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 14.161798548415648, | |
| "grad_norm": 28.316547393798828, | |
| "learning_rate": 6.486890501760391e-07, | |
| "loss": 0.253, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 14.161798548415648, | |
| "eval_accuracy": 0.8313253012048193, | |
| "eval_f1": 0.8302605210420841, | |
| "eval_loss": 0.3548405170440674, | |
| "eval_precision": 0.8510682004930156, | |
| "eval_recall": 0.8104460093896714, | |
| "eval_runtime": 48.5444, | |
| "eval_samples_per_second": 206.883, | |
| "eval_steps_per_second": 12.937, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 14.338821030270845, | |
| "grad_norm": 16.459693908691406, | |
| "learning_rate": 6.290198855254617e-07, | |
| "loss": 0.2522, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 14.338821030270845, | |
| "eval_accuracy": 0.8271432838793189, | |
| "eval_f1": 0.8302698474775128, | |
| "eval_loss": 0.3594595193862915, | |
| "eval_precision": 0.8299452697419859, | |
| "eval_recall": 0.8305946791862285, | |
| "eval_runtime": 48.5687, | |
| "eval_samples_per_second": 206.779, | |
| "eval_steps_per_second": 12.93, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 14.51584351212604, | |
| "grad_norm": 54.176578521728516, | |
| "learning_rate": 6.093507208748843e-07, | |
| "loss": 0.2418, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 14.51584351212604, | |
| "eval_accuracy": 0.8300308672707358, | |
| "eval_f1": 0.8312407315867524, | |
| "eval_loss": 0.3571302890777588, | |
| "eval_precision": 0.8402958225064961, | |
| "eval_recall": 0.8223787167449139, | |
| "eval_runtime": 48.4796, | |
| "eval_samples_per_second": 207.159, | |
| "eval_steps_per_second": 12.954, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 14.692865993981236, | |
| "grad_norm": 18.803722381591797, | |
| "learning_rate": 5.896815562243072e-07, | |
| "loss": 0.2543, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 14.692865993981236, | |
| "eval_accuracy": 0.8324205914567361, | |
| "eval_f1": 0.8432523051131601, | |
| "eval_loss": 0.36417821049690247, | |
| "eval_precision": 0.8048, | |
| "eval_recall": 0.8855633802816901, | |
| "eval_runtime": 48.459, | |
| "eval_samples_per_second": 207.247, | |
| "eval_steps_per_second": 12.959, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 14.869888475836431, | |
| "grad_norm": 6.331387996673584, | |
| "learning_rate": 5.700123915737298e-07, | |
| "loss": 0.2483, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 14.869888475836431, | |
| "eval_accuracy": 0.828636861495569, | |
| "eval_f1": 0.83049345021176, | |
| "eval_loss": 0.3517729938030243, | |
| "eval_precision": 0.8363419956357866, | |
| "eval_recall": 0.8247261345852895, | |
| "eval_runtime": 48.6656, | |
| "eval_samples_per_second": 206.368, | |
| "eval_steps_per_second": 12.904, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 15.046910957691626, | |
| "grad_norm": 63.694122314453125, | |
| "learning_rate": 5.503432269231526e-07, | |
| "loss": 0.2489, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 15.046910957691626, | |
| "eval_accuracy": 0.8317235885691526, | |
| "eval_f1": 0.8441247002398081, | |
| "eval_loss": 0.35860148072242737, | |
| "eval_precision": 0.7986038394415358, | |
| "eval_recall": 0.8951486697965572, | |
| "eval_runtime": 48.5388, | |
| "eval_samples_per_second": 206.907, | |
| "eval_steps_per_second": 12.938, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 15.223933439546823, | |
| "grad_norm": 18.07522201538086, | |
| "learning_rate": 5.306740622725753e-07, | |
| "loss": 0.2439, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 15.223933439546823, | |
| "eval_accuracy": 0.8292342925420692, | |
| "eval_f1": 0.8383143207315924, | |
| "eval_loss": 0.3605121970176697, | |
| "eval_precision": 0.8090991810737034, | |
| "eval_recall": 0.8697183098591549, | |
| "eval_runtime": 48.5578, | |
| "eval_samples_per_second": 206.826, | |
| "eval_steps_per_second": 12.933, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 15.400955921402018, | |
| "grad_norm": 6.01281213760376, | |
| "learning_rate": 5.110048976219979e-07, | |
| "loss": 0.2463, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 15.400955921402018, | |
| "eval_accuracy": 0.8317235885691526, | |
| "eval_f1": 0.8435764531654942, | |
| "eval_loss": 0.3613792955875397, | |
| "eval_precision": 0.800597329585383, | |
| "eval_recall": 0.8914319248826291, | |
| "eval_runtime": 48.4585, | |
| "eval_samples_per_second": 207.25, | |
| "eval_steps_per_second": 12.96, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 15.577978403257214, | |
| "grad_norm": 4.72086763381958, | |
| "learning_rate": 4.913357329714207e-07, | |
| "loss": 0.2421, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 15.577978403257214, | |
| "eval_accuracy": 0.8301304391118192, | |
| "eval_f1": 0.8313896026882783, | |
| "eval_loss": 0.3474978506565094, | |
| "eval_precision": 0.8401917698761486, | |
| "eval_recall": 0.8227699530516432, | |
| "eval_runtime": 48.4981, | |
| "eval_samples_per_second": 207.08, | |
| "eval_steps_per_second": 12.949, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 15.755000885112409, | |
| "grad_norm": 7.4664530754089355, | |
| "learning_rate": 4.716665683208434e-07, | |
| "loss": 0.2449, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 15.755000885112409, | |
| "eval_accuracy": 0.8310265856815693, | |
| "eval_f1": 0.8286723876829883, | |
| "eval_loss": 0.35074958205223083, | |
| "eval_precision": 0.856248696015022, | |
| "eval_recall": 0.8028169014084507, | |
| "eval_runtime": 48.4074, | |
| "eval_samples_per_second": 207.468, | |
| "eval_steps_per_second": 12.973, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 15.932023366967606, | |
| "grad_norm": 18.542203903198242, | |
| "learning_rate": 4.5199740367026614e-07, | |
| "loss": 0.242, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 15.932023366967606, | |
| "eval_accuracy": 0.8300308672707358, | |
| "eval_f1": 0.8335121427874769, | |
| "eval_loss": 0.34495285153388977, | |
| "eval_precision": 0.831161252674577, | |
| "eval_recall": 0.8358763693270735, | |
| "eval_runtime": 48.5317, | |
| "eval_samples_per_second": 206.937, | |
| "eval_steps_per_second": 12.94, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 16.1090458488228, | |
| "grad_norm": 53.64085006713867, | |
| "learning_rate": 4.323282390196888e-07, | |
| "loss": 0.2477, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 16.1090458488228, | |
| "eval_accuracy": 0.8307278701583193, | |
| "eval_f1": 0.8348552554886342, | |
| "eval_loss": 0.3650229275226593, | |
| "eval_precision": 0.8292165187186414, | |
| "eval_recall": 0.8405712050078247, | |
| "eval_runtime": 48.4566, | |
| "eval_samples_per_second": 207.257, | |
| "eval_steps_per_second": 12.96, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 16.286068330677995, | |
| "grad_norm": 10.626422882080078, | |
| "learning_rate": 4.126590743691115e-07, | |
| "loss": 0.2388, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 16.286068330677995, | |
| "eval_accuracy": 0.8311261575226526, | |
| "eval_f1": 0.8387526145655068, | |
| "eval_loss": 0.35832735896110535, | |
| "eval_precision": 0.8159452460229375, | |
| "eval_recall": 0.8628716744913928, | |
| "eval_runtime": 48.551, | |
| "eval_samples_per_second": 206.855, | |
| "eval_steps_per_second": 12.935, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 16.463090812533192, | |
| "grad_norm": 23.993255615234375, | |
| "learning_rate": 3.9298990971853425e-07, | |
| "loss": 0.2469, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 16.463090812533192, | |
| "eval_accuracy": 0.8298317235885692, | |
| "eval_f1": 0.8339648304673078, | |
| "eval_loss": 0.3484230637550354, | |
| "eval_precision": 0.8284115035707392, | |
| "eval_recall": 0.8395931142410016, | |
| "eval_runtime": 48.4135, | |
| "eval_samples_per_second": 207.442, | |
| "eval_steps_per_second": 12.972, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 16.64011329438839, | |
| "grad_norm": 3.7660863399505615, | |
| "learning_rate": 3.73320745067957e-07, | |
| "loss": 0.235, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 16.64011329438839, | |
| "eval_accuracy": 0.8298317235885692, | |
| "eval_f1": 0.8365375418460067, | |
| "eval_loss": 0.3643123507499695, | |
| "eval_precision": 0.8184540520306943, | |
| "eval_recall": 0.8554381846635368, | |
| "eval_runtime": 48.5272, | |
| "eval_samples_per_second": 206.956, | |
| "eval_steps_per_second": 12.941, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 16.817135776243582, | |
| "grad_norm": 17.013097763061523, | |
| "learning_rate": 3.536515804173796e-07, | |
| "loss": 0.2336, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 16.817135776243582, | |
| "eval_accuracy": 0.827740714925819, | |
| "eval_f1": 0.8336858296481446, | |
| "eval_loss": 0.34748417139053345, | |
| "eval_precision": 0.8196597353497165, | |
| "eval_recall": 0.8482003129890454, | |
| "eval_runtime": 48.3887, | |
| "eval_samples_per_second": 207.548, | |
| "eval_steps_per_second": 12.978, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 16.99415825809878, | |
| "grad_norm": 17.022958755493164, | |
| "learning_rate": 3.3398241576680236e-07, | |
| "loss": 0.2455, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 16.99415825809878, | |
| "eval_accuracy": 0.8288360051777357, | |
| "eval_f1": 0.8308903098868667, | |
| "eval_loss": 0.348213791847229, | |
| "eval_precision": 0.8357411438749258, | |
| "eval_recall": 0.826095461658842, | |
| "eval_runtime": 48.5355, | |
| "eval_samples_per_second": 206.921, | |
| "eval_steps_per_second": 12.939, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 17.171180739953975, | |
| "grad_norm": 6.458636283874512, | |
| "learning_rate": 3.143132511162251e-07, | |
| "loss": 0.2372, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 17.171180739953975, | |
| "eval_accuracy": 0.836503036941153, | |
| "eval_f1": 0.8473693995166388, | |
| "eval_loss": 0.35692235827445984, | |
| "eval_precision": 0.8072972015586256, | |
| "eval_recall": 0.8916275430359938, | |
| "eval_runtime": 48.4702, | |
| "eval_samples_per_second": 207.2, | |
| "eval_steps_per_second": 12.956, | |
| "step": 97000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 112980, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.9198903946321024e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |