|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 19.995409685563462, |
|
"eval_steps": 500, |
|
"global_step": 21780, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.09180628873077806, |
|
"grad_norm": 0.25305476784706116, |
|
"learning_rate": 0.0005, |
|
"loss": 1.6083, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.18361257746155613, |
|
"grad_norm": 0.3124815821647644, |
|
"learning_rate": 0.0005, |
|
"loss": 1.5865, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.2754188661923342, |
|
"grad_norm": 0.33532094955444336, |
|
"learning_rate": 0.0005, |
|
"loss": 1.5663, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.36722515492311225, |
|
"grad_norm": 0.3140541911125183, |
|
"learning_rate": 0.0005, |
|
"loss": 1.5709, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.4590314436538903, |
|
"grad_norm": 0.3288426399230957, |
|
"learning_rate": 0.0005, |
|
"loss": 1.567, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.5508377323846684, |
|
"grad_norm": 0.362998902797699, |
|
"learning_rate": 0.0005, |
|
"loss": 1.5635, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.6426440211154464, |
|
"grad_norm": 0.37498417496681213, |
|
"learning_rate": 0.0005, |
|
"loss": 1.5651, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.7344503098462245, |
|
"grad_norm": 0.45461997389793396, |
|
"learning_rate": 0.0005, |
|
"loss": 1.5687, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.8262565985770025, |
|
"grad_norm": 0.42744582891464233, |
|
"learning_rate": 0.0005, |
|
"loss": 1.5349, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 0.9180628873077806, |
|
"grad_norm": 0.43451881408691406, |
|
"learning_rate": 0.0005, |
|
"loss": 1.5463, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.999770484278173, |
|
"eval_accuracy": 0.6871615720524017, |
|
"eval_loss": 1.3538845777511597, |
|
"eval_runtime": 9.1369, |
|
"eval_samples_per_second": 54.723, |
|
"eval_steps_per_second": 6.895, |
|
"step": 1089 |
|
}, |
|
{ |
|
"epoch": 1.0098691760385587, |
|
"grad_norm": 0.45076310634613037, |
|
"learning_rate": 0.0005, |
|
"loss": 1.51, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 1.1016754647693368, |
|
"grad_norm": 0.47397756576538086, |
|
"learning_rate": 0.0005, |
|
"loss": 1.2876, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 1.1934817535001148, |
|
"grad_norm": 0.4858494997024536, |
|
"learning_rate": 0.0005, |
|
"loss": 1.2932, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 1.2852880422308928, |
|
"grad_norm": 0.52638179063797, |
|
"learning_rate": 0.0005, |
|
"loss": 1.2933, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 1.377094330961671, |
|
"grad_norm": 0.5119564533233643, |
|
"learning_rate": 0.0005, |
|
"loss": 1.2981, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 1.468900619692449, |
|
"grad_norm": 0.5230050086975098, |
|
"learning_rate": 0.0005, |
|
"loss": 1.32, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 1.560706908423227, |
|
"grad_norm": 0.5544503331184387, |
|
"learning_rate": 0.0005, |
|
"loss": 1.2968, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 1.652513197154005, |
|
"grad_norm": 0.6211464405059814, |
|
"learning_rate": 0.0005, |
|
"loss": 1.3169, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 1.744319485884783, |
|
"grad_norm": 0.5929824709892273, |
|
"learning_rate": 0.0005, |
|
"loss": 1.3062, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 1.836125774615561, |
|
"grad_norm": 0.559938371181488, |
|
"learning_rate": 0.0005, |
|
"loss": 1.3056, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 1.9279320633463393, |
|
"grad_norm": 0.6206551194190979, |
|
"learning_rate": 0.0005, |
|
"loss": 1.3199, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 1.999540968556346, |
|
"eval_accuracy": 0.7021979621542941, |
|
"eval_loss": 1.1631555557250977, |
|
"eval_runtime": 9.1947, |
|
"eval_samples_per_second": 54.379, |
|
"eval_steps_per_second": 6.852, |
|
"step": 2178 |
|
}, |
|
{ |
|
"epoch": 2.0197383520771175, |
|
"grad_norm": 0.5680427551269531, |
|
"learning_rate": 0.0005, |
|
"loss": 1.2504, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 2.1115446408078955, |
|
"grad_norm": 0.6365711688995361, |
|
"learning_rate": 0.0005, |
|
"loss": 0.9978, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 2.2033509295386735, |
|
"grad_norm": 0.7458738088607788, |
|
"learning_rate": 0.0005, |
|
"loss": 1.0328, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 2.2951572182694515, |
|
"grad_norm": 0.5955780148506165, |
|
"learning_rate": 0.0005, |
|
"loss": 1.0487, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 2.3869635070002295, |
|
"grad_norm": 0.6577792763710022, |
|
"learning_rate": 0.0005, |
|
"loss": 1.0702, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 2.4787697957310075, |
|
"grad_norm": 0.6620481014251709, |
|
"learning_rate": 0.0005, |
|
"loss": 1.0653, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 2.5705760844617855, |
|
"grad_norm": 0.7260810136795044, |
|
"learning_rate": 0.0005, |
|
"loss": 1.0881, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 2.6623823731925635, |
|
"grad_norm": 0.7388891577720642, |
|
"learning_rate": 0.0005, |
|
"loss": 1.0823, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 2.754188661923342, |
|
"grad_norm": 0.7343183159828186, |
|
"learning_rate": 0.0005, |
|
"loss": 1.1044, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 2.84599495065412, |
|
"grad_norm": 0.606133759021759, |
|
"learning_rate": 0.0005, |
|
"loss": 1.0884, |
|
"step": 3100 |
|
}, |
|
{ |
|
"epoch": 2.937801239384898, |
|
"grad_norm": 0.6684651374816895, |
|
"learning_rate": 0.0005, |
|
"loss": 1.1039, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 2.9993114528345193, |
|
"eval_accuracy": 0.713353711790393, |
|
"eval_loss": 1.0347490310668945, |
|
"eval_runtime": 9.0955, |
|
"eval_samples_per_second": 54.972, |
|
"eval_steps_per_second": 6.927, |
|
"step": 3267 |
|
}, |
|
{ |
|
"epoch": 3.029607528115676, |
|
"grad_norm": 0.7068032026290894, |
|
"learning_rate": 0.0005, |
|
"loss": 1.0062, |
|
"step": 3300 |
|
}, |
|
{ |
|
"epoch": 3.121413816846454, |
|
"grad_norm": 0.7648475766181946, |
|
"learning_rate": 0.0005, |
|
"loss": 0.8266, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 3.213220105577232, |
|
"grad_norm": 0.8599951267242432, |
|
"learning_rate": 0.0005, |
|
"loss": 0.8518, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 3.30502639430801, |
|
"grad_norm": 0.8535987138748169, |
|
"learning_rate": 0.0005, |
|
"loss": 0.8849, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 3.396832683038788, |
|
"grad_norm": 0.8338284492492676, |
|
"learning_rate": 0.0005, |
|
"loss": 0.8896, |
|
"step": 3700 |
|
}, |
|
{ |
|
"epoch": 3.488638971769566, |
|
"grad_norm": 0.8858312964439392, |
|
"learning_rate": 0.0005, |
|
"loss": 0.9036, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 3.580445260500344, |
|
"grad_norm": 0.8100723028182983, |
|
"learning_rate": 0.0005, |
|
"loss": 0.9088, |
|
"step": 3900 |
|
}, |
|
{ |
|
"epoch": 3.672251549231122, |
|
"grad_norm": 0.827248215675354, |
|
"learning_rate": 0.0005, |
|
"loss": 0.9252, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 3.7640578379619005, |
|
"grad_norm": 0.7961249351501465, |
|
"learning_rate": 0.0005, |
|
"loss": 0.9277, |
|
"step": 4100 |
|
}, |
|
{ |
|
"epoch": 3.8558641266926785, |
|
"grad_norm": 0.8417057394981384, |
|
"learning_rate": 0.0005, |
|
"loss": 0.9406, |
|
"step": 4200 |
|
}, |
|
{ |
|
"epoch": 3.9476704154234565, |
|
"grad_norm": 0.8393632173538208, |
|
"learning_rate": 0.0005, |
|
"loss": 0.9356, |
|
"step": 4300 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.723665211062591, |
|
"eval_loss": 0.9234349131584167, |
|
"eval_runtime": 9.127, |
|
"eval_samples_per_second": 54.783, |
|
"eval_steps_per_second": 6.903, |
|
"step": 4357 |
|
}, |
|
{ |
|
"epoch": 4.039476704154235, |
|
"grad_norm": 0.7149174809455872, |
|
"learning_rate": 0.0005, |
|
"loss": 0.8365, |
|
"step": 4400 |
|
}, |
|
{ |
|
"epoch": 4.131282992885013, |
|
"grad_norm": 0.7990247011184692, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6924, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 4.223089281615791, |
|
"grad_norm": 0.7871200442314148, |
|
"learning_rate": 0.0005, |
|
"loss": 0.7195, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 4.314895570346569, |
|
"grad_norm": 0.8242298364639282, |
|
"learning_rate": 0.0005, |
|
"loss": 0.749, |
|
"step": 4700 |
|
}, |
|
{ |
|
"epoch": 4.406701859077347, |
|
"grad_norm": 0.864937424659729, |
|
"learning_rate": 0.0005, |
|
"loss": 0.7616, |
|
"step": 4800 |
|
}, |
|
{ |
|
"epoch": 4.498508147808125, |
|
"grad_norm": 0.8644908666610718, |
|
"learning_rate": 0.0005, |
|
"loss": 0.7763, |
|
"step": 4900 |
|
}, |
|
{ |
|
"epoch": 4.590314436538903, |
|
"grad_norm": 0.9300737977027893, |
|
"learning_rate": 0.0005, |
|
"loss": 0.7999, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 4.682120725269681, |
|
"grad_norm": 0.8608301877975464, |
|
"learning_rate": 0.0005, |
|
"loss": 0.8112, |
|
"step": 5100 |
|
}, |
|
{ |
|
"epoch": 4.773927014000459, |
|
"grad_norm": 0.8489959239959717, |
|
"learning_rate": 0.0005, |
|
"loss": 0.8168, |
|
"step": 5200 |
|
}, |
|
{ |
|
"epoch": 4.865733302731237, |
|
"grad_norm": 0.9379121661186218, |
|
"learning_rate": 0.0005, |
|
"loss": 0.8256, |
|
"step": 5300 |
|
}, |
|
{ |
|
"epoch": 4.957539591462015, |
|
"grad_norm": 0.9303742051124573, |
|
"learning_rate": 0.0005, |
|
"loss": 0.8312, |
|
"step": 5400 |
|
}, |
|
{ |
|
"epoch": 4.999770484278173, |
|
"eval_accuracy": 0.7306957787481805, |
|
"eval_loss": 0.8529494404792786, |
|
"eval_runtime": 9.1302, |
|
"eval_samples_per_second": 54.764, |
|
"eval_steps_per_second": 6.9, |
|
"step": 5446 |
|
}, |
|
{ |
|
"epoch": 5.049345880192793, |
|
"grad_norm": 0.8555835485458374, |
|
"learning_rate": 0.0005, |
|
"loss": 0.7084, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 5.141152168923571, |
|
"grad_norm": 0.99271559715271, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6162, |
|
"step": 5600 |
|
}, |
|
{ |
|
"epoch": 5.232958457654349, |
|
"grad_norm": 0.9135051965713501, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6335, |
|
"step": 5700 |
|
}, |
|
{ |
|
"epoch": 5.324764746385127, |
|
"grad_norm": 0.9890821576118469, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6725, |
|
"step": 5800 |
|
}, |
|
{ |
|
"epoch": 5.416571035115905, |
|
"grad_norm": 0.941581130027771, |
|
"learning_rate": 0.0005, |
|
"loss": 0.677, |
|
"step": 5900 |
|
}, |
|
{ |
|
"epoch": 5.508377323846684, |
|
"grad_norm": 0.8984777927398682, |
|
"learning_rate": 0.0005, |
|
"loss": 0.704, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 5.600183612577462, |
|
"grad_norm": 1.007198452949524, |
|
"learning_rate": 0.0005, |
|
"loss": 0.71, |
|
"step": 6100 |
|
}, |
|
{ |
|
"epoch": 5.69198990130824, |
|
"grad_norm": 1.0259089469909668, |
|
"learning_rate": 0.0005, |
|
"loss": 0.731, |
|
"step": 6200 |
|
}, |
|
{ |
|
"epoch": 5.783796190039018, |
|
"grad_norm": 1.0021580457687378, |
|
"learning_rate": 0.0005, |
|
"loss": 0.7425, |
|
"step": 6300 |
|
}, |
|
{ |
|
"epoch": 5.875602478769796, |
|
"grad_norm": 0.9952398538589478, |
|
"learning_rate": 0.0005, |
|
"loss": 0.7395, |
|
"step": 6400 |
|
}, |
|
{ |
|
"epoch": 5.967408767500574, |
|
"grad_norm": 1.257237195968628, |
|
"learning_rate": 0.0005, |
|
"loss": 0.7565, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 5.999540968556346, |
|
"eval_accuracy": 0.7372139737991267, |
|
"eval_loss": 0.7859806418418884, |
|
"eval_runtime": 9.1037, |
|
"eval_samples_per_second": 54.923, |
|
"eval_steps_per_second": 6.92, |
|
"step": 6535 |
|
}, |
|
{ |
|
"epoch": 6.059215056231352, |
|
"grad_norm": 0.9014455676078796, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6172, |
|
"step": 6600 |
|
}, |
|
{ |
|
"epoch": 6.15102134496213, |
|
"grad_norm": 0.8993759751319885, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5578, |
|
"step": 6700 |
|
}, |
|
{ |
|
"epoch": 6.242827633692908, |
|
"grad_norm": 0.9781313538551331, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5846, |
|
"step": 6800 |
|
}, |
|
{ |
|
"epoch": 6.334633922423686, |
|
"grad_norm": 0.9638663530349731, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6045, |
|
"step": 6900 |
|
}, |
|
{ |
|
"epoch": 6.426440211154464, |
|
"grad_norm": 0.9154132604598999, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6203, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 6.518246499885242, |
|
"grad_norm": 1.0229175090789795, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6421, |
|
"step": 7100 |
|
}, |
|
{ |
|
"epoch": 6.61005278861602, |
|
"grad_norm": 1.0271860361099243, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6525, |
|
"step": 7200 |
|
}, |
|
{ |
|
"epoch": 6.701859077346798, |
|
"grad_norm": 0.981940746307373, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6625, |
|
"step": 7300 |
|
}, |
|
{ |
|
"epoch": 6.793665366077576, |
|
"grad_norm": 1.0368313789367676, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6754, |
|
"step": 7400 |
|
}, |
|
{ |
|
"epoch": 6.885471654808354, |
|
"grad_norm": 1.058945655822754, |
|
"learning_rate": 0.0005, |
|
"loss": 0.685, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 6.977277943539132, |
|
"grad_norm": 0.9555875062942505, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6985, |
|
"step": 7600 |
|
}, |
|
{ |
|
"epoch": 6.999311452834519, |
|
"eval_accuracy": 0.741475982532751, |
|
"eval_loss": 0.7415383458137512, |
|
"eval_runtime": 9.1507, |
|
"eval_samples_per_second": 54.641, |
|
"eval_steps_per_second": 6.885, |
|
"step": 7624 |
|
}, |
|
{ |
|
"epoch": 7.06908423226991, |
|
"grad_norm": 0.9311098456382751, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5515, |
|
"step": 7700 |
|
}, |
|
{ |
|
"epoch": 7.160890521000688, |
|
"grad_norm": 0.9383052587509155, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5276, |
|
"step": 7800 |
|
}, |
|
{ |
|
"epoch": 7.252696809731467, |
|
"grad_norm": 0.9980023503303528, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5369, |
|
"step": 7900 |
|
}, |
|
{ |
|
"epoch": 7.344503098462245, |
|
"grad_norm": 0.9550295472145081, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5676, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 7.436309387193023, |
|
"grad_norm": 0.9989355206489563, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5805, |
|
"step": 8100 |
|
}, |
|
{ |
|
"epoch": 7.528115675923801, |
|
"grad_norm": 1.0541164875030518, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6059, |
|
"step": 8200 |
|
}, |
|
{ |
|
"epoch": 7.619921964654579, |
|
"grad_norm": 1.1819795370101929, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6102, |
|
"step": 8300 |
|
}, |
|
{ |
|
"epoch": 7.711728253385357, |
|
"grad_norm": 1.0276963710784912, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6196, |
|
"step": 8400 |
|
}, |
|
{ |
|
"epoch": 7.803534542116135, |
|
"grad_norm": 0.9429015517234802, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6251, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 7.895340830846913, |
|
"grad_norm": 1.1022661924362183, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6355, |
|
"step": 8600 |
|
}, |
|
{ |
|
"epoch": 7.987147119577691, |
|
"grad_norm": 0.9936983585357666, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6623, |
|
"step": 8700 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.7457467248908297, |
|
"eval_loss": 0.7111318111419678, |
|
"eval_runtime": 9.2592, |
|
"eval_samples_per_second": 54.0, |
|
"eval_steps_per_second": 6.804, |
|
"step": 8714 |
|
}, |
|
{ |
|
"epoch": 8.07895340830847, |
|
"grad_norm": 0.9469543099403381, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5012, |
|
"step": 8800 |
|
}, |
|
{ |
|
"epoch": 8.170759697039248, |
|
"grad_norm": 1.0641309022903442, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4854, |
|
"step": 8900 |
|
}, |
|
{ |
|
"epoch": 8.262565985770026, |
|
"grad_norm": 1.0741512775421143, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5116, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 8.354372274500804, |
|
"grad_norm": 1.038459300994873, |
|
"learning_rate": 0.0005, |
|
"loss": 0.532, |
|
"step": 9100 |
|
}, |
|
{ |
|
"epoch": 8.446178563231582, |
|
"grad_norm": 0.976216197013855, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5478, |
|
"step": 9200 |
|
}, |
|
{ |
|
"epoch": 8.53798485196236, |
|
"grad_norm": 1.0488590002059937, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5649, |
|
"step": 9300 |
|
}, |
|
{ |
|
"epoch": 8.629791140693138, |
|
"grad_norm": 1.0634474754333496, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5873, |
|
"step": 9400 |
|
}, |
|
{ |
|
"epoch": 8.721597429423916, |
|
"grad_norm": 1.0678510665893555, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5926, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 8.813403718154694, |
|
"grad_norm": 1.0706602334976196, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6089, |
|
"step": 9600 |
|
}, |
|
{ |
|
"epoch": 8.905210006885472, |
|
"grad_norm": 1.0411357879638672, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6096, |
|
"step": 9700 |
|
}, |
|
{ |
|
"epoch": 8.99701629561625, |
|
"grad_norm": 1.0408014059066772, |
|
"learning_rate": 0.0005, |
|
"loss": 0.6281, |
|
"step": 9800 |
|
}, |
|
{ |
|
"epoch": 8.999770484278173, |
|
"eval_accuracy": 0.7481106259097525, |
|
"eval_loss": 0.6775466203689575, |
|
"eval_runtime": 9.131, |
|
"eval_samples_per_second": 54.759, |
|
"eval_steps_per_second": 6.9, |
|
"step": 9803 |
|
}, |
|
{ |
|
"epoch": 9.088822584347028, |
|
"grad_norm": 0.9965059757232666, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4564, |
|
"step": 9900 |
|
}, |
|
{ |
|
"epoch": 9.180628873077806, |
|
"grad_norm": 0.9503998160362244, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4683, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 9.272435161808584, |
|
"grad_norm": 0.9788022041320801, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4948, |
|
"step": 10100 |
|
}, |
|
{ |
|
"epoch": 9.364241450539362, |
|
"grad_norm": 0.97702556848526, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5083, |
|
"step": 10200 |
|
}, |
|
{ |
|
"epoch": 9.45604773927014, |
|
"grad_norm": 1.0470614433288574, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5269, |
|
"step": 10300 |
|
}, |
|
{ |
|
"epoch": 9.547854028000918, |
|
"grad_norm": 1.1560702323913574, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5454, |
|
"step": 10400 |
|
}, |
|
{ |
|
"epoch": 9.639660316731696, |
|
"grad_norm": 1.065807819366455, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5582, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 9.731466605462474, |
|
"grad_norm": 1.0357719659805298, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5673, |
|
"step": 10600 |
|
}, |
|
{ |
|
"epoch": 9.823272894193252, |
|
"grad_norm": 1.051635503768921, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5808, |
|
"step": 10700 |
|
}, |
|
{ |
|
"epoch": 9.91507918292403, |
|
"grad_norm": 1.1748321056365967, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5885, |
|
"step": 10800 |
|
}, |
|
{ |
|
"epoch": 9.999540968556346, |
|
"eval_accuracy": 0.7495545851528385, |
|
"eval_loss": 0.668917715549469, |
|
"eval_runtime": 9.0726, |
|
"eval_samples_per_second": 55.111, |
|
"eval_steps_per_second": 6.944, |
|
"step": 10892 |
|
}, |
|
{ |
|
"epoch": 10.006885471654808, |
|
"grad_norm": 1.015367031097412, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5866, |
|
"step": 10900 |
|
}, |
|
{ |
|
"epoch": 10.098691760385586, |
|
"grad_norm": 1.0127650499343872, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4348, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 10.190498049116364, |
|
"grad_norm": 0.9805082082748413, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4539, |
|
"step": 11100 |
|
}, |
|
{ |
|
"epoch": 10.282304337847142, |
|
"grad_norm": 1.007502555847168, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4772, |
|
"step": 11200 |
|
}, |
|
{ |
|
"epoch": 10.37411062657792, |
|
"grad_norm": 1.1225751638412476, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4984, |
|
"step": 11300 |
|
}, |
|
{ |
|
"epoch": 10.465916915308698, |
|
"grad_norm": 1.164787769317627, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5082, |
|
"step": 11400 |
|
}, |
|
{ |
|
"epoch": 10.557723204039476, |
|
"grad_norm": 1.1108081340789795, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5247, |
|
"step": 11500 |
|
}, |
|
{ |
|
"epoch": 10.649529492770254, |
|
"grad_norm": 1.0103906393051147, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5299, |
|
"step": 11600 |
|
}, |
|
{ |
|
"epoch": 10.741335781501032, |
|
"grad_norm": 1.0926460027694702, |
|
"learning_rate": 0.0005, |
|
"loss": 0.553, |
|
"step": 11700 |
|
}, |
|
{ |
|
"epoch": 10.83314207023181, |
|
"grad_norm": 1.1170562505722046, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5623, |
|
"step": 11800 |
|
}, |
|
{ |
|
"epoch": 10.924948358962588, |
|
"grad_norm": 0.9009856581687927, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5721, |
|
"step": 11900 |
|
}, |
|
{ |
|
"epoch": 10.999311452834519, |
|
"eval_accuracy": 0.7530101892285298, |
|
"eval_loss": 0.6363697648048401, |
|
"eval_runtime": 9.1706, |
|
"eval_samples_per_second": 54.522, |
|
"eval_steps_per_second": 6.87, |
|
"step": 11981 |
|
}, |
|
{ |
|
"epoch": 11.016754647693366, |
|
"grad_norm": 0.8974876403808594, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5493, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 11.108560936424144, |
|
"grad_norm": 0.974591851234436, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4203, |
|
"step": 12100 |
|
}, |
|
{ |
|
"epoch": 11.200367225154924, |
|
"grad_norm": 1.0000176429748535, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4454, |
|
"step": 12200 |
|
}, |
|
{ |
|
"epoch": 11.292173513885702, |
|
"grad_norm": 1.0163604021072388, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4579, |
|
"step": 12300 |
|
}, |
|
{ |
|
"epoch": 11.38397980261648, |
|
"grad_norm": 1.0325731039047241, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4852, |
|
"step": 12400 |
|
}, |
|
{ |
|
"epoch": 11.475786091347258, |
|
"grad_norm": 1.093885064125061, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4979, |
|
"step": 12500 |
|
}, |
|
{ |
|
"epoch": 11.567592380078036, |
|
"grad_norm": 1.0193454027175903, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5076, |
|
"step": 12600 |
|
}, |
|
{ |
|
"epoch": 11.659398668808814, |
|
"grad_norm": 0.9466171264648438, |
|
"learning_rate": 0.0005, |
|
"loss": 0.529, |
|
"step": 12700 |
|
}, |
|
{ |
|
"epoch": 11.751204957539592, |
|
"grad_norm": 1.1748363971710205, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5398, |
|
"step": 12800 |
|
}, |
|
{ |
|
"epoch": 11.84301124627037, |
|
"grad_norm": 1.0464725494384766, |
|
"learning_rate": 0.0005, |
|
"loss": 0.544, |
|
"step": 12900 |
|
}, |
|
{ |
|
"epoch": 11.934817535001148, |
|
"grad_norm": 1.1592164039611816, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5504, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_accuracy": 0.7540669577874818, |
|
"eval_loss": 0.6318646669387817, |
|
"eval_runtime": 9.1786, |
|
"eval_samples_per_second": 54.474, |
|
"eval_steps_per_second": 6.864, |
|
"step": 13071 |
|
}, |
|
{ |
|
"epoch": 12.026623823731926, |
|
"grad_norm": 0.935130774974823, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5135, |
|
"step": 13100 |
|
}, |
|
{ |
|
"epoch": 12.118430112462704, |
|
"grad_norm": 1.0784605741500854, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4078, |
|
"step": 13200 |
|
}, |
|
{ |
|
"epoch": 12.210236401193482, |
|
"grad_norm": 1.0809942483901978, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4313, |
|
"step": 13300 |
|
}, |
|
{ |
|
"epoch": 12.30204268992426, |
|
"grad_norm": 1.1434837579727173, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4576, |
|
"step": 13400 |
|
}, |
|
{ |
|
"epoch": 12.393848978655038, |
|
"grad_norm": 1.0381824970245361, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4688, |
|
"step": 13500 |
|
}, |
|
{ |
|
"epoch": 12.485655267385816, |
|
"grad_norm": 1.0997393131256104, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4838, |
|
"step": 13600 |
|
}, |
|
{ |
|
"epoch": 12.577461556116594, |
|
"grad_norm": 1.1423929929733276, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5015, |
|
"step": 13700 |
|
}, |
|
{ |
|
"epoch": 12.669267844847372, |
|
"grad_norm": 1.0119812488555908, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5099, |
|
"step": 13800 |
|
}, |
|
{ |
|
"epoch": 12.76107413357815, |
|
"grad_norm": 1.0487439632415771, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5224, |
|
"step": 13900 |
|
}, |
|
{ |
|
"epoch": 12.852880422308928, |
|
"grad_norm": 1.1909295320510864, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5362, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 12.944686711039706, |
|
"grad_norm": 1.0172119140625, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5406, |
|
"step": 14100 |
|
}, |
|
{ |
|
"epoch": 12.999770484278173, |
|
"eval_accuracy": 0.7549374090247453, |
|
"eval_loss": 0.6185360550880432, |
|
"eval_runtime": 9.1246, |
|
"eval_samples_per_second": 54.797, |
|
"eval_steps_per_second": 6.904, |
|
"step": 14160 |
|
}, |
|
{ |
|
"epoch": 13.036492999770484, |
|
"grad_norm": 0.8250077366828918, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4829, |
|
"step": 14200 |
|
}, |
|
{ |
|
"epoch": 13.128299288501262, |
|
"grad_norm": 0.9478662014007568, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4008, |
|
"step": 14300 |
|
}, |
|
{ |
|
"epoch": 13.22010557723204, |
|
"grad_norm": 0.9901746511459351, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4221, |
|
"step": 14400 |
|
}, |
|
{ |
|
"epoch": 13.311911865962818, |
|
"grad_norm": 1.0091050863265991, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4457, |
|
"step": 14500 |
|
}, |
|
{ |
|
"epoch": 13.403718154693596, |
|
"grad_norm": 1.1044436693191528, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4612, |
|
"step": 14600 |
|
}, |
|
{ |
|
"epoch": 13.495524443424374, |
|
"grad_norm": 1.095940113067627, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4808, |
|
"step": 14700 |
|
}, |
|
{ |
|
"epoch": 13.587330732155152, |
|
"grad_norm": 0.9664422869682312, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4936, |
|
"step": 14800 |
|
}, |
|
{ |
|
"epoch": 13.67913702088593, |
|
"grad_norm": 0.9911946654319763, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5029, |
|
"step": 14900 |
|
}, |
|
{ |
|
"epoch": 13.770943309616708, |
|
"grad_norm": 1.0246057510375977, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5111, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 13.862749598347486, |
|
"grad_norm": 1.094871163368225, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5253, |
|
"step": 15100 |
|
}, |
|
{ |
|
"epoch": 13.954555887078264, |
|
"grad_norm": 1.1010475158691406, |
|
"learning_rate": 0.0005, |
|
"loss": 0.536, |
|
"step": 15200 |
|
}, |
|
{ |
|
"epoch": 13.999540968556346, |
|
"eval_accuracy": 0.7565123726346433, |
|
"eval_loss": 0.61576908826828, |
|
"eval_runtime": 9.1949, |
|
"eval_samples_per_second": 54.378, |
|
"eval_steps_per_second": 6.852, |
|
"step": 15249 |
|
}, |
|
{ |
|
"epoch": 14.046362175809042, |
|
"grad_norm": 0.9305674433708191, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4565, |
|
"step": 15300 |
|
}, |
|
{ |
|
"epoch": 14.13816846453982, |
|
"grad_norm": 0.9582420587539673, |
|
"learning_rate": 0.0005, |
|
"loss": 0.3955, |
|
"step": 15400 |
|
}, |
|
{ |
|
"epoch": 14.229974753270598, |
|
"grad_norm": 1.0967265367507935, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4182, |
|
"step": 15500 |
|
}, |
|
{ |
|
"epoch": 14.321781042001376, |
|
"grad_norm": 0.9896946549415588, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4357, |
|
"step": 15600 |
|
}, |
|
{ |
|
"epoch": 14.413587330732156, |
|
"grad_norm": 1.0576809644699097, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4549, |
|
"step": 15700 |
|
}, |
|
{ |
|
"epoch": 14.505393619462934, |
|
"grad_norm": 1.0679750442504883, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4694, |
|
"step": 15800 |
|
}, |
|
{ |
|
"epoch": 14.597199908193712, |
|
"grad_norm": 1.0860179662704468, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4857, |
|
"step": 15900 |
|
}, |
|
{ |
|
"epoch": 14.68900619692449, |
|
"grad_norm": 1.0913867950439453, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4935, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 14.780812485655268, |
|
"grad_norm": 1.0317550897598267, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5047, |
|
"step": 16100 |
|
}, |
|
{ |
|
"epoch": 14.872618774386046, |
|
"grad_norm": 1.0847545862197876, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5103, |
|
"step": 16200 |
|
}, |
|
{ |
|
"epoch": 14.964425063116824, |
|
"grad_norm": 1.0414918661117554, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5205, |
|
"step": 16300 |
|
}, |
|
{ |
|
"epoch": 14.999311452834519, |
|
"eval_accuracy": 0.757839883551674, |
|
"eval_loss": 0.5975773930549622, |
|
"eval_runtime": 9.2897, |
|
"eval_samples_per_second": 53.823, |
|
"eval_steps_per_second": 6.782, |
|
"step": 16338 |
|
}, |
|
{ |
|
"epoch": 15.056231351847602, |
|
"grad_norm": 0.8716037273406982, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4352, |
|
"step": 16400 |
|
}, |
|
{ |
|
"epoch": 15.14803764057838, |
|
"grad_norm": 0.9753520488739014, |
|
"learning_rate": 0.0005, |
|
"loss": 0.3923, |
|
"step": 16500 |
|
}, |
|
{ |
|
"epoch": 15.239843929309158, |
|
"grad_norm": 1.0317981243133545, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4083, |
|
"step": 16600 |
|
}, |
|
{ |
|
"epoch": 15.331650218039936, |
|
"grad_norm": 1.0025997161865234, |
|
"learning_rate": 0.0005, |
|
"loss": 0.435, |
|
"step": 16700 |
|
}, |
|
{ |
|
"epoch": 15.423456506770714, |
|
"grad_norm": 1.0460282564163208, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4479, |
|
"step": 16800 |
|
}, |
|
{ |
|
"epoch": 15.515262795501492, |
|
"grad_norm": 0.957950234413147, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4613, |
|
"step": 16900 |
|
}, |
|
{ |
|
"epoch": 15.60706908423227, |
|
"grad_norm": 1.0739928483963013, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4724, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 15.698875372963048, |
|
"grad_norm": 1.161801815032959, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4889, |
|
"step": 17100 |
|
}, |
|
{ |
|
"epoch": 15.790681661693826, |
|
"grad_norm": 1.1013407707214355, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4911, |
|
"step": 17200 |
|
}, |
|
{ |
|
"epoch": 15.882487950424604, |
|
"grad_norm": 1.1333537101745605, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5082, |
|
"step": 17300 |
|
}, |
|
{ |
|
"epoch": 15.974294239155382, |
|
"grad_norm": 1.1869903802871704, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5175, |
|
"step": 17400 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_accuracy": 0.7590393013100437, |
|
"eval_loss": 0.5922391414642334, |
|
"eval_runtime": 9.1385, |
|
"eval_samples_per_second": 54.714, |
|
"eval_steps_per_second": 6.894, |
|
"step": 17428 |
|
}, |
|
{ |
|
"epoch": 16.06610052788616, |
|
"grad_norm": 1.069451093673706, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4122, |
|
"step": 17500 |
|
}, |
|
{ |
|
"epoch": 16.15790681661694, |
|
"grad_norm": 0.9926499724388123, |
|
"learning_rate": 0.0005, |
|
"loss": 0.3909, |
|
"step": 17600 |
|
}, |
|
{ |
|
"epoch": 16.249713105347716, |
|
"grad_norm": 1.067694902420044, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4075, |
|
"step": 17700 |
|
}, |
|
{ |
|
"epoch": 16.341519394078496, |
|
"grad_norm": 0.9541024565696716, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4215, |
|
"step": 17800 |
|
}, |
|
{ |
|
"epoch": 16.433325682809272, |
|
"grad_norm": 1.0835317373275757, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4405, |
|
"step": 17900 |
|
}, |
|
{ |
|
"epoch": 16.525131971540052, |
|
"grad_norm": 1.1008630990982056, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4562, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 16.616938260270828, |
|
"grad_norm": 1.1548954248428345, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4694, |
|
"step": 18100 |
|
}, |
|
{ |
|
"epoch": 16.708744549001608, |
|
"grad_norm": 0.9584476947784424, |
|
"learning_rate": 0.0005, |
|
"loss": 0.481, |
|
"step": 18200 |
|
}, |
|
{ |
|
"epoch": 16.800550837732384, |
|
"grad_norm": 1.0248852968215942, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4854, |
|
"step": 18300 |
|
}, |
|
{ |
|
"epoch": 16.892357126463164, |
|
"grad_norm": 1.1495635509490967, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5012, |
|
"step": 18400 |
|
}, |
|
{ |
|
"epoch": 16.98416341519394, |
|
"grad_norm": 1.0732232332229614, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5068, |
|
"step": 18500 |
|
}, |
|
{ |
|
"epoch": 16.999770484278173, |
|
"eval_accuracy": 0.759278020378457, |
|
"eval_loss": 0.582272469997406, |
|
"eval_runtime": 9.1627, |
|
"eval_samples_per_second": 54.569, |
|
"eval_steps_per_second": 6.876, |
|
"step": 18517 |
|
}, |
|
{ |
|
"epoch": 17.07596970392472, |
|
"grad_norm": 1.0219801664352417, |
|
"learning_rate": 0.0005, |
|
"loss": 0.3925, |
|
"step": 18600 |
|
}, |
|
{ |
|
"epoch": 17.167775992655496, |
|
"grad_norm": 0.8995177149772644, |
|
"learning_rate": 0.0005, |
|
"loss": 0.384, |
|
"step": 18700 |
|
}, |
|
{ |
|
"epoch": 17.259582281386276, |
|
"grad_norm": 1.0561939477920532, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4041, |
|
"step": 18800 |
|
}, |
|
{ |
|
"epoch": 17.351388570117052, |
|
"grad_norm": 0.9551050662994385, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4246, |
|
"step": 18900 |
|
}, |
|
{ |
|
"epoch": 17.443194858847832, |
|
"grad_norm": 1.0561895370483398, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4382, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 17.53500114757861, |
|
"grad_norm": 1.1141507625579834, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4583, |
|
"step": 19100 |
|
}, |
|
{ |
|
"epoch": 17.626807436309388, |
|
"grad_norm": 1.141851544380188, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4609, |
|
"step": 19200 |
|
}, |
|
{ |
|
"epoch": 17.718613725040164, |
|
"grad_norm": 1.1055680513381958, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4668, |
|
"step": 19300 |
|
}, |
|
{ |
|
"epoch": 17.810420013770944, |
|
"grad_norm": 1.0966593027114868, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4843, |
|
"step": 19400 |
|
}, |
|
{ |
|
"epoch": 17.90222630250172, |
|
"grad_norm": 1.1193351745605469, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4933, |
|
"step": 19500 |
|
}, |
|
{ |
|
"epoch": 17.9940325912325, |
|
"grad_norm": 1.258386254310608, |
|
"learning_rate": 0.0005, |
|
"loss": 0.5023, |
|
"step": 19600 |
|
}, |
|
{ |
|
"epoch": 17.999540968556346, |
|
"eval_accuracy": 0.760660844250364, |
|
"eval_loss": 0.5753906965255737, |
|
"eval_runtime": 9.1571, |
|
"eval_samples_per_second": 54.603, |
|
"eval_steps_per_second": 6.88, |
|
"step": 19606 |
|
}, |
|
{ |
|
"epoch": 18.085838879963276, |
|
"grad_norm": 0.9986138939857483, |
|
"learning_rate": 0.0005, |
|
"loss": 0.3719, |
|
"step": 19700 |
|
}, |
|
{ |
|
"epoch": 18.177645168694056, |
|
"grad_norm": 0.9715769290924072, |
|
"learning_rate": 0.0005, |
|
"loss": 0.3825, |
|
"step": 19800 |
|
}, |
|
{ |
|
"epoch": 18.269451457424832, |
|
"grad_norm": 1.0640110969543457, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4056, |
|
"step": 19900 |
|
}, |
|
{ |
|
"epoch": 18.361257746155612, |
|
"grad_norm": 1.0777605772018433, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4172, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 18.45306403488639, |
|
"grad_norm": 1.0681456327438354, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4318, |
|
"step": 20100 |
|
}, |
|
{ |
|
"epoch": 18.544870323617168, |
|
"grad_norm": 1.1611887216567993, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4431, |
|
"step": 20200 |
|
}, |
|
{ |
|
"epoch": 18.636676612347944, |
|
"grad_norm": 0.9606027603149414, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4594, |
|
"step": 20300 |
|
}, |
|
{ |
|
"epoch": 18.728482901078724, |
|
"grad_norm": 1.0940121412277222, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4692, |
|
"step": 20400 |
|
}, |
|
{ |
|
"epoch": 18.8202891898095, |
|
"grad_norm": 1.219136357307434, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4734, |
|
"step": 20500 |
|
}, |
|
{ |
|
"epoch": 18.91209547854028, |
|
"grad_norm": 1.058712363243103, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4848, |
|
"step": 20600 |
|
}, |
|
{ |
|
"epoch": 18.99931145283452, |
|
"eval_accuracy": 0.7608238719068413, |
|
"eval_loss": 0.5781408548355103, |
|
"eval_runtime": 9.1618, |
|
"eval_samples_per_second": 54.574, |
|
"eval_steps_per_second": 6.876, |
|
"step": 20695 |
|
}, |
|
{ |
|
"epoch": 19.003901767271056, |
|
"grad_norm": 0.853932797908783, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4916, |
|
"step": 20700 |
|
}, |
|
{ |
|
"epoch": 19.095708056001836, |
|
"grad_norm": 1.0715103149414062, |
|
"learning_rate": 0.0005, |
|
"loss": 0.362, |
|
"step": 20800 |
|
}, |
|
{ |
|
"epoch": 19.187514344732612, |
|
"grad_norm": 0.9922650456428528, |
|
"learning_rate": 0.0005, |
|
"loss": 0.3766, |
|
"step": 20900 |
|
}, |
|
{ |
|
"epoch": 19.279320633463392, |
|
"grad_norm": 1.0666494369506836, |
|
"learning_rate": 0.0005, |
|
"loss": 0.3999, |
|
"step": 21000 |
|
}, |
|
{ |
|
"epoch": 19.371126922194172, |
|
"grad_norm": 1.026061773300171, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4188, |
|
"step": 21100 |
|
}, |
|
{ |
|
"epoch": 19.462933210924948, |
|
"grad_norm": 1.0832582712173462, |
|
"learning_rate": 0.0005, |
|
"loss": 0.432, |
|
"step": 21200 |
|
}, |
|
{ |
|
"epoch": 19.554739499655728, |
|
"grad_norm": 1.109492540359497, |
|
"learning_rate": 0.0005, |
|
"loss": 0.436, |
|
"step": 21300 |
|
}, |
|
{ |
|
"epoch": 19.646545788386504, |
|
"grad_norm": 1.0509155988693237, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4564, |
|
"step": 21400 |
|
}, |
|
{ |
|
"epoch": 19.738352077117284, |
|
"grad_norm": 1.1579090356826782, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4622, |
|
"step": 21500 |
|
}, |
|
{ |
|
"epoch": 19.83015836584806, |
|
"grad_norm": 1.0362952947616577, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4728, |
|
"step": 21600 |
|
}, |
|
{ |
|
"epoch": 19.92196465457884, |
|
"grad_norm": 1.06185781955719, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4767, |
|
"step": 21700 |
|
}, |
|
{ |
|
"epoch": 19.995409685563462, |
|
"eval_accuracy": 0.7611848617176128, |
|
"eval_loss": 0.5757241249084473, |
|
"eval_runtime": 9.1042, |
|
"eval_samples_per_second": 54.92, |
|
"eval_steps_per_second": 6.92, |
|
"step": 21780 |
|
}, |
|
{ |
|
"epoch": 19.995409685563462, |
|
"step": 21780, |
|
"total_flos": 2.2953223726028554e+18, |
|
"train_loss": 0.6653460843285034, |
|
"train_runtime": 50472.5887, |
|
"train_samples_per_second": 13.811, |
|
"train_steps_per_second": 0.432 |
|
} |
|
], |
|
"logging_steps": 100, |
|
"max_steps": 21780, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 20, |
|
"save_steps": 500, |
|
"total_flos": 2.2953223726028554e+18, |
|
"train_batch_size": 1, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|