| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 50, | |
| "global_step": 3819, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03927729772191673, | |
| "grad_norm": 0.3003779351711273, | |
| "learning_rate": 6.544502617801048e-05, | |
| "loss": 4.2186, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03927729772191673, | |
| "eval_loss": 3.861752510070801, | |
| "eval_runtime": 29.4492, | |
| "eval_samples_per_second": 3.396, | |
| "eval_steps_per_second": 0.441, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07855459544383346, | |
| "grad_norm": 0.19962768256664276, | |
| "learning_rate": 0.00013089005235602096, | |
| "loss": 3.1772, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07855459544383346, | |
| "eval_loss": 2.59340238571167, | |
| "eval_runtime": 29.4336, | |
| "eval_samples_per_second": 3.397, | |
| "eval_steps_per_second": 0.442, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1178318931657502, | |
| "grad_norm": 0.17451965808868408, | |
| "learning_rate": 0.00019633507853403142, | |
| "loss": 2.401, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.1178318931657502, | |
| "eval_loss": 2.2559566497802734, | |
| "eval_runtime": 29.4433, | |
| "eval_samples_per_second": 3.396, | |
| "eval_steps_per_second": 0.442, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.15710919088766692, | |
| "grad_norm": 0.18111871182918549, | |
| "learning_rate": 0.0002617801047120419, | |
| "loss": 2.1397, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15710919088766692, | |
| "eval_loss": 2.1369433403015137, | |
| "eval_runtime": 29.6429, | |
| "eval_samples_per_second": 3.373, | |
| "eval_steps_per_second": 0.439, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.19638648860958366, | |
| "grad_norm": 0.18973998725414276, | |
| "learning_rate": 0.0003272251308900523, | |
| "loss": 2.0834, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.19638648860958366, | |
| "eval_loss": 2.08046555519104, | |
| "eval_runtime": 29.5575, | |
| "eval_samples_per_second": 3.383, | |
| "eval_steps_per_second": 0.44, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.2356637863315004, | |
| "grad_norm": 0.1734466403722763, | |
| "learning_rate": 0.00039267015706806284, | |
| "loss": 2.055, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.2356637863315004, | |
| "eval_loss": 2.056304693222046, | |
| "eval_runtime": 29.4932, | |
| "eval_samples_per_second": 3.391, | |
| "eval_steps_per_second": 0.441, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.2749410840534171, | |
| "grad_norm": 0.21972551941871643, | |
| "learning_rate": 0.0004581151832460733, | |
| "loss": 2.043, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.2749410840534171, | |
| "eval_loss": 2.0286335945129395, | |
| "eval_runtime": 29.3134, | |
| "eval_samples_per_second": 3.411, | |
| "eval_steps_per_second": 0.443, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.31421838177533384, | |
| "grad_norm": 0.20237264037132263, | |
| "learning_rate": 0.000497381437299971, | |
| "loss": 2.0135, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.31421838177533384, | |
| "eval_loss": 2.0177054405212402, | |
| "eval_runtime": 29.516, | |
| "eval_samples_per_second": 3.388, | |
| "eval_steps_per_second": 0.44, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.3534956794972506, | |
| "grad_norm": 0.18964040279388428, | |
| "learning_rate": 0.0004901076520221123, | |
| "loss": 1.9971, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.3534956794972506, | |
| "eval_loss": 2.0019736289978027, | |
| "eval_runtime": 29.578, | |
| "eval_samples_per_second": 3.381, | |
| "eval_steps_per_second": 0.44, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.3927729772191673, | |
| "grad_norm": 0.20875422656536102, | |
| "learning_rate": 0.00048283386674425374, | |
| "loss": 1.9766, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3927729772191673, | |
| "eval_loss": 1.991439700126648, | |
| "eval_runtime": 29.3577, | |
| "eval_samples_per_second": 3.406, | |
| "eval_steps_per_second": 0.443, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.43205027494108406, | |
| "grad_norm": 0.1806900054216385, | |
| "learning_rate": 0.00047556008146639513, | |
| "loss": 1.9677, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.43205027494108406, | |
| "eval_loss": 1.9789185523986816, | |
| "eval_runtime": 29.5883, | |
| "eval_samples_per_second": 3.38, | |
| "eval_steps_per_second": 0.439, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.4713275726630008, | |
| "grad_norm": 0.17965032160282135, | |
| "learning_rate": 0.00046828629618853653, | |
| "loss": 1.9562, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.4713275726630008, | |
| "eval_loss": 1.9679882526397705, | |
| "eval_runtime": 29.7136, | |
| "eval_samples_per_second": 3.365, | |
| "eval_steps_per_second": 0.438, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.5106048703849175, | |
| "grad_norm": 0.16700418293476105, | |
| "learning_rate": 0.0004610125109106779, | |
| "loss": 1.9594, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.5106048703849175, | |
| "eval_loss": 1.9631274938583374, | |
| "eval_runtime": 29.8472, | |
| "eval_samples_per_second": 3.35, | |
| "eval_steps_per_second": 0.436, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.5498821681068342, | |
| "grad_norm": 0.1479049026966095, | |
| "learning_rate": 0.0004537387256328193, | |
| "loss": 1.9423, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.5498821681068342, | |
| "eval_loss": 1.9546252489089966, | |
| "eval_runtime": 29.5017, | |
| "eval_samples_per_second": 3.39, | |
| "eval_steps_per_second": 0.441, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.589159465828751, | |
| "grad_norm": 0.14641450345516205, | |
| "learning_rate": 0.0004464649403549607, | |
| "loss": 1.9587, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.589159465828751, | |
| "eval_loss": 1.9469780921936035, | |
| "eval_runtime": 29.8753, | |
| "eval_samples_per_second": 3.347, | |
| "eval_steps_per_second": 0.435, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.6284367635506677, | |
| "grad_norm": 0.19608749449253082, | |
| "learning_rate": 0.0004391911550771021, | |
| "loss": 1.9408, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.6284367635506677, | |
| "eval_loss": 1.9397352933883667, | |
| "eval_runtime": 29.4107, | |
| "eval_samples_per_second": 3.4, | |
| "eval_steps_per_second": 0.442, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.6677140612725845, | |
| "grad_norm": 0.18671496212482452, | |
| "learning_rate": 0.0004319173697992435, | |
| "loss": 1.9816, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.6677140612725845, | |
| "eval_loss": 1.9425079822540283, | |
| "eval_runtime": 29.4898, | |
| "eval_samples_per_second": 3.391, | |
| "eval_steps_per_second": 0.441, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.7069913589945012, | |
| "grad_norm": 0.16066303849220276, | |
| "learning_rate": 0.00042464358452138495, | |
| "loss": 1.9298, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.7069913589945012, | |
| "eval_loss": 1.9176784753799438, | |
| "eval_runtime": 29.544, | |
| "eval_samples_per_second": 3.385, | |
| "eval_steps_per_second": 0.44, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.746268656716418, | |
| "grad_norm": 0.17130565643310547, | |
| "learning_rate": 0.00041736979924352635, | |
| "loss": 1.9021, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.746268656716418, | |
| "eval_loss": 1.9149932861328125, | |
| "eval_runtime": 29.4076, | |
| "eval_samples_per_second": 3.4, | |
| "eval_steps_per_second": 0.442, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.7855459544383346, | |
| "grad_norm": 0.15130634605884552, | |
| "learning_rate": 0.00041009601396566775, | |
| "loss": 1.9104, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7855459544383346, | |
| "eval_loss": 1.9071648120880127, | |
| "eval_runtime": 29.4527, | |
| "eval_samples_per_second": 3.395, | |
| "eval_steps_per_second": 0.441, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8248232521602514, | |
| "grad_norm": 0.1651030331850052, | |
| "learning_rate": 0.00040282222868780914, | |
| "loss": 1.9325, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.8248232521602514, | |
| "eval_loss": 1.8993133306503296, | |
| "eval_runtime": 29.5512, | |
| "eval_samples_per_second": 3.384, | |
| "eval_steps_per_second": 0.44, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.8641005498821681, | |
| "grad_norm": 0.16182482242584229, | |
| "learning_rate": 0.00039554844340995054, | |
| "loss": 1.9183, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8641005498821681, | |
| "eval_loss": 1.9054021835327148, | |
| "eval_runtime": 29.3876, | |
| "eval_samples_per_second": 3.403, | |
| "eval_steps_per_second": 0.442, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.9033778476040848, | |
| "grad_norm": 0.16718855500221252, | |
| "learning_rate": 0.00038827465813209193, | |
| "loss": 1.9557, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.9033778476040848, | |
| "eval_loss": 1.894838571548462, | |
| "eval_runtime": 29.428, | |
| "eval_samples_per_second": 3.398, | |
| "eval_steps_per_second": 0.442, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.9426551453260016, | |
| "grad_norm": 0.14458701014518738, | |
| "learning_rate": 0.0003810008728542333, | |
| "loss": 1.9261, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.9426551453260016, | |
| "eval_loss": 1.8822929859161377, | |
| "eval_runtime": 29.4891, | |
| "eval_samples_per_second": 3.391, | |
| "eval_steps_per_second": 0.441, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.9819324430479183, | |
| "grad_norm": 0.1621202528476715, | |
| "learning_rate": 0.0003737270875763747, | |
| "loss": 1.9337, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.9819324430479183, | |
| "eval_loss": 1.8784922361373901, | |
| "eval_runtime": 29.4313, | |
| "eval_samples_per_second": 3.398, | |
| "eval_steps_per_second": 0.442, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.021209740769835, | |
| "grad_norm": 0.17562049627304077, | |
| "learning_rate": 0.00036645330229851617, | |
| "loss": 1.9034, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.021209740769835, | |
| "eval_loss": 1.8769853115081787, | |
| "eval_runtime": 29.4816, | |
| "eval_samples_per_second": 3.392, | |
| "eval_steps_per_second": 0.441, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.0604870384917517, | |
| "grad_norm": 0.18526038527488708, | |
| "learning_rate": 0.00035917951702065757, | |
| "loss": 1.8603, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.0604870384917517, | |
| "eval_loss": 1.8668268918991089, | |
| "eval_runtime": 29.5392, | |
| "eval_samples_per_second": 3.385, | |
| "eval_steps_per_second": 0.44, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.0997643362136684, | |
| "grad_norm": 0.1711549311876297, | |
| "learning_rate": 0.00035190573174279896, | |
| "loss": 1.8477, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.0997643362136684, | |
| "eval_loss": 1.8661786317825317, | |
| "eval_runtime": 29.4511, | |
| "eval_samples_per_second": 3.395, | |
| "eval_steps_per_second": 0.441, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.139041633935585, | |
| "grad_norm": 0.18905235826969147, | |
| "learning_rate": 0.00034463194646494036, | |
| "loss": 1.8658, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.139041633935585, | |
| "eval_loss": 1.8573988676071167, | |
| "eval_runtime": 29.4384, | |
| "eval_samples_per_second": 3.397, | |
| "eval_steps_per_second": 0.442, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.178318931657502, | |
| "grad_norm": 0.16769160330295563, | |
| "learning_rate": 0.0003373581611870818, | |
| "loss": 1.8923, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.178318931657502, | |
| "eval_loss": 1.857424259185791, | |
| "eval_runtime": 29.4845, | |
| "eval_samples_per_second": 3.392, | |
| "eval_steps_per_second": 0.441, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.2175962293794187, | |
| "grad_norm": 0.2095043659210205, | |
| "learning_rate": 0.0003300843759092232, | |
| "loss": 1.8777, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.2175962293794187, | |
| "eval_loss": 1.8602547645568848, | |
| "eval_runtime": 29.5084, | |
| "eval_samples_per_second": 3.389, | |
| "eval_steps_per_second": 0.441, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.2568735271013354, | |
| "grad_norm": 0.18356919288635254, | |
| "learning_rate": 0.0003228105906313646, | |
| "loss": 1.8645, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.2568735271013354, | |
| "eval_loss": 1.8516952991485596, | |
| "eval_runtime": 29.539, | |
| "eval_samples_per_second": 3.385, | |
| "eval_steps_per_second": 0.44, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.2961508248232523, | |
| "grad_norm": 0.23483967781066895, | |
| "learning_rate": 0.000315536805353506, | |
| "loss": 1.8204, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.2961508248232523, | |
| "eval_loss": 1.8447459936141968, | |
| "eval_runtime": 29.4621, | |
| "eval_samples_per_second": 3.394, | |
| "eval_steps_per_second": 0.441, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.335428122545169, | |
| "grad_norm": 0.21005484461784363, | |
| "learning_rate": 0.0003082630200756474, | |
| "loss": 1.8661, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.335428122545169, | |
| "eval_loss": 1.8399604558944702, | |
| "eval_runtime": 29.4709, | |
| "eval_samples_per_second": 3.393, | |
| "eval_steps_per_second": 0.441, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.3747054202670856, | |
| "grad_norm": 0.19819919764995575, | |
| "learning_rate": 0.0003009892347977888, | |
| "loss": 1.8595, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.3747054202670856, | |
| "eval_loss": 1.8384497165679932, | |
| "eval_runtime": 29.4803, | |
| "eval_samples_per_second": 3.392, | |
| "eval_steps_per_second": 0.441, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.4139827179890023, | |
| "grad_norm": 0.21313180029392242, | |
| "learning_rate": 0.0002937154495199302, | |
| "loss": 1.857, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.4139827179890023, | |
| "eval_loss": 1.831393837928772, | |
| "eval_runtime": 29.4965, | |
| "eval_samples_per_second": 3.39, | |
| "eval_steps_per_second": 0.441, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.453260015710919, | |
| "grad_norm": 0.2181958109140396, | |
| "learning_rate": 0.0002864416642420716, | |
| "loss": 1.8431, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.453260015710919, | |
| "eval_loss": 1.827854037284851, | |
| "eval_runtime": 29.4535, | |
| "eval_samples_per_second": 3.395, | |
| "eval_steps_per_second": 0.441, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.4925373134328357, | |
| "grad_norm": 0.2101767212152481, | |
| "learning_rate": 0.00027916787896421297, | |
| "loss": 1.8249, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.4925373134328357, | |
| "eval_loss": 1.8285325765609741, | |
| "eval_runtime": 29.5818, | |
| "eval_samples_per_second": 3.38, | |
| "eval_steps_per_second": 0.439, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.5318146111547526, | |
| "grad_norm": 0.23812232911586761, | |
| "learning_rate": 0.00027189409368635437, | |
| "loss": 1.8372, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.5318146111547526, | |
| "eval_loss": 1.824296474456787, | |
| "eval_runtime": 29.4834, | |
| "eval_samples_per_second": 3.392, | |
| "eval_steps_per_second": 0.441, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.5710919088766693, | |
| "grad_norm": 0.1917562186717987, | |
| "learning_rate": 0.00026462030840849576, | |
| "loss": 1.8589, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.5710919088766693, | |
| "eval_loss": 1.8210265636444092, | |
| "eval_runtime": 29.5979, | |
| "eval_samples_per_second": 3.379, | |
| "eval_steps_per_second": 0.439, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.6103692065985862, | |
| "grad_norm": 0.222442165017128, | |
| "learning_rate": 0.00025734652313063716, | |
| "loss": 1.829, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.6103692065985862, | |
| "eval_loss": 1.8052531480789185, | |
| "eval_runtime": 29.6699, | |
| "eval_samples_per_second": 3.37, | |
| "eval_steps_per_second": 0.438, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.6496465043205029, | |
| "grad_norm": 0.19593429565429688, | |
| "learning_rate": 0.00025007273785277855, | |
| "loss": 1.8154, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6496465043205029, | |
| "eval_loss": 1.800155758857727, | |
| "eval_runtime": 29.4349, | |
| "eval_samples_per_second": 3.397, | |
| "eval_steps_per_second": 0.442, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6889238020424195, | |
| "grad_norm": 0.22215288877487183, | |
| "learning_rate": 0.00024279895257492, | |
| "loss": 1.8122, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.6889238020424195, | |
| "eval_loss": 1.8008465766906738, | |
| "eval_runtime": 29.3948, | |
| "eval_samples_per_second": 3.402, | |
| "eval_steps_per_second": 0.442, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.7282010997643362, | |
| "grad_norm": 0.20354895293712616, | |
| "learning_rate": 0.0002355251672970614, | |
| "loss": 1.8297, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.7282010997643362, | |
| "eval_loss": 1.7969157695770264, | |
| "eval_runtime": 29.4741, | |
| "eval_samples_per_second": 3.393, | |
| "eval_steps_per_second": 0.441, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.767478397486253, | |
| "grad_norm": 0.2116580605506897, | |
| "learning_rate": 0.0002282513820192028, | |
| "loss": 1.8467, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.767478397486253, | |
| "eval_loss": 1.796259880065918, | |
| "eval_runtime": 29.3661, | |
| "eval_samples_per_second": 3.405, | |
| "eval_steps_per_second": 0.443, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.8067556952081696, | |
| "grad_norm": 0.19466279447078705, | |
| "learning_rate": 0.00022097759674134421, | |
| "loss": 1.8242, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.8067556952081696, | |
| "eval_loss": 1.797297716140747, | |
| "eval_runtime": 29.4909, | |
| "eval_samples_per_second": 3.391, | |
| "eval_steps_per_second": 0.441, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.8460329929300863, | |
| "grad_norm": 0.20243284106254578, | |
| "learning_rate": 0.0002137038114634856, | |
| "loss": 1.8209, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.8460329929300863, | |
| "eval_loss": 1.7902172803878784, | |
| "eval_runtime": 29.5667, | |
| "eval_samples_per_second": 3.382, | |
| "eval_steps_per_second": 0.44, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.8853102906520032, | |
| "grad_norm": 0.20039430260658264, | |
| "learning_rate": 0.000206430026185627, | |
| "loss": 1.8193, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.8853102906520032, | |
| "eval_loss": 1.7889854907989502, | |
| "eval_runtime": 29.5198, | |
| "eval_samples_per_second": 3.388, | |
| "eval_steps_per_second": 0.44, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.9245875883739199, | |
| "grad_norm": 0.1968667209148407, | |
| "learning_rate": 0.0001991562409077684, | |
| "loss": 1.8153, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.9245875883739199, | |
| "eval_loss": 1.7838914394378662, | |
| "eval_runtime": 29.4874, | |
| "eval_samples_per_second": 3.391, | |
| "eval_steps_per_second": 0.441, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.9638648860958368, | |
| "grad_norm": 0.2244279533624649, | |
| "learning_rate": 0.0001918824556299098, | |
| "loss": 1.7845, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.9638648860958368, | |
| "eval_loss": 1.7779937982559204, | |
| "eval_runtime": 29.5562, | |
| "eval_samples_per_second": 3.383, | |
| "eval_steps_per_second": 0.44, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0031421838177534, | |
| "grad_norm": 0.20871788263320923, | |
| "learning_rate": 0.0001846086703520512, | |
| "loss": 1.7975, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 2.0031421838177534, | |
| "eval_loss": 1.779373288154602, | |
| "eval_runtime": 29.5808, | |
| "eval_samples_per_second": 3.381, | |
| "eval_steps_per_second": 0.439, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 2.04241948153967, | |
| "grad_norm": 0.2194184958934784, | |
| "learning_rate": 0.0001773348850741926, | |
| "loss": 1.7922, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.04241948153967, | |
| "eval_loss": 1.7732738256454468, | |
| "eval_runtime": 29.4165, | |
| "eval_samples_per_second": 3.399, | |
| "eval_steps_per_second": 0.442, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.081696779261587, | |
| "grad_norm": 0.21199199557304382, | |
| "learning_rate": 0.00017006109979633404, | |
| "loss": 1.7558, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 2.081696779261587, | |
| "eval_loss": 1.772108554840088, | |
| "eval_runtime": 29.5171, | |
| "eval_samples_per_second": 3.388, | |
| "eval_steps_per_second": 0.44, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 2.1209740769835035, | |
| "grad_norm": 0.2122960090637207, | |
| "learning_rate": 0.00016278731451847543, | |
| "loss": 1.7821, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.1209740769835035, | |
| "eval_loss": 1.7694271802902222, | |
| "eval_runtime": 29.9036, | |
| "eval_samples_per_second": 3.344, | |
| "eval_steps_per_second": 0.435, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.16025137470542, | |
| "grad_norm": 0.2310086041688919, | |
| "learning_rate": 0.00015551352924061683, | |
| "loss": 1.7735, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.16025137470542, | |
| "eval_loss": 1.7644474506378174, | |
| "eval_runtime": 29.5003, | |
| "eval_samples_per_second": 3.39, | |
| "eval_steps_per_second": 0.441, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.199528672427337, | |
| "grad_norm": 0.23034410178661346, | |
| "learning_rate": 0.00014823974396275822, | |
| "loss": 1.7802, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.199528672427337, | |
| "eval_loss": 1.7630077600479126, | |
| "eval_runtime": 29.3837, | |
| "eval_samples_per_second": 3.403, | |
| "eval_steps_per_second": 0.442, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.2388059701492535, | |
| "grad_norm": 0.2436445653438568, | |
| "learning_rate": 0.00014096595868489962, | |
| "loss": 1.7616, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 2.2388059701492535, | |
| "eval_loss": 1.7602554559707642, | |
| "eval_runtime": 29.4586, | |
| "eval_samples_per_second": 3.395, | |
| "eval_steps_per_second": 0.441, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 2.27808326787117, | |
| "grad_norm": 0.24777990579605103, | |
| "learning_rate": 0.00013369217340704104, | |
| "loss": 1.7751, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.27808326787117, | |
| "eval_loss": 1.7579619884490967, | |
| "eval_runtime": 29.3692, | |
| "eval_samples_per_second": 3.405, | |
| "eval_steps_per_second": 0.443, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.3173605655930873, | |
| "grad_norm": 0.21132971346378326, | |
| "learning_rate": 0.00012641838812918243, | |
| "loss": 1.7811, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 2.3173605655930873, | |
| "eval_loss": 1.754958987236023, | |
| "eval_runtime": 29.4284, | |
| "eval_samples_per_second": 3.398, | |
| "eval_steps_per_second": 0.442, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 2.356637863315004, | |
| "grad_norm": 0.2826303541660309, | |
| "learning_rate": 0.00011914460285132383, | |
| "loss": 1.7356, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.356637863315004, | |
| "eval_loss": 1.7528626918792725, | |
| "eval_runtime": 29.4272, | |
| "eval_samples_per_second": 3.398, | |
| "eval_steps_per_second": 0.442, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.3959151610369207, | |
| "grad_norm": 0.20650461316108704, | |
| "learning_rate": 0.00011187081757346524, | |
| "loss": 1.7575, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 2.3959151610369207, | |
| "eval_loss": 1.751355767250061, | |
| "eval_runtime": 29.577, | |
| "eval_samples_per_second": 3.381, | |
| "eval_steps_per_second": 0.44, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 2.4351924587588374, | |
| "grad_norm": 0.22746974229812622, | |
| "learning_rate": 0.00010459703229560663, | |
| "loss": 1.7547, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.4351924587588374, | |
| "eval_loss": 1.7509515285491943, | |
| "eval_runtime": 29.3774, | |
| "eval_samples_per_second": 3.404, | |
| "eval_steps_per_second": 0.443, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.474469756480754, | |
| "grad_norm": 0.23667258024215698, | |
| "learning_rate": 9.732324701774804e-05, | |
| "loss": 1.7699, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 2.474469756480754, | |
| "eval_loss": 1.752171516418457, | |
| "eval_runtime": 29.4311, | |
| "eval_samples_per_second": 3.398, | |
| "eval_steps_per_second": 0.442, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 2.5137470542026707, | |
| "grad_norm": 0.25468263030052185, | |
| "learning_rate": 9.004946173988944e-05, | |
| "loss": 1.7506, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.5137470542026707, | |
| "eval_loss": 1.7496337890625, | |
| "eval_runtime": 29.4486, | |
| "eval_samples_per_second": 3.396, | |
| "eval_steps_per_second": 0.441, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.5530243519245874, | |
| "grad_norm": 0.24980412423610687, | |
| "learning_rate": 8.277567646203083e-05, | |
| "loss": 1.7564, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 2.5530243519245874, | |
| "eval_loss": 1.7441329956054688, | |
| "eval_runtime": 29.3681, | |
| "eval_samples_per_second": 3.405, | |
| "eval_steps_per_second": 0.443, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 2.5923016496465046, | |
| "grad_norm": 0.2186252772808075, | |
| "learning_rate": 7.550189118417226e-05, | |
| "loss": 1.7517, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.5923016496465046, | |
| "eval_loss": 1.7436232566833496, | |
| "eval_runtime": 29.5214, | |
| "eval_samples_per_second": 3.387, | |
| "eval_steps_per_second": 0.44, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "grad_norm": 0.21136987209320068, | |
| "learning_rate": 6.822810590631365e-05, | |
| "loss": 1.7371, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "eval_loss": 1.7433184385299683, | |
| "eval_runtime": 29.6962, | |
| "eval_samples_per_second": 3.367, | |
| "eval_steps_per_second": 0.438, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 2.670856245090338, | |
| "grad_norm": 0.2606451213359833, | |
| "learning_rate": 6.095432062845505e-05, | |
| "loss": 1.7425, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.670856245090338, | |
| "eval_loss": 1.742987036705017, | |
| "eval_runtime": 29.4333, | |
| "eval_samples_per_second": 3.398, | |
| "eval_steps_per_second": 0.442, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.7101335428122546, | |
| "grad_norm": 0.218011736869812, | |
| "learning_rate": 5.368053535059645e-05, | |
| "loss": 1.7407, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 2.7101335428122546, | |
| "eval_loss": 1.7402327060699463, | |
| "eval_runtime": 29.6244, | |
| "eval_samples_per_second": 3.376, | |
| "eval_steps_per_second": 0.439, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 2.7494108405341713, | |
| "grad_norm": 0.2643437385559082, | |
| "learning_rate": 4.640675007273786e-05, | |
| "loss": 1.7513, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.7494108405341713, | |
| "eval_loss": 1.740808129310608, | |
| "eval_runtime": 29.3534, | |
| "eval_samples_per_second": 3.407, | |
| "eval_steps_per_second": 0.443, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.788688138256088, | |
| "grad_norm": 0.26832252740859985, | |
| "learning_rate": 3.913296479487925e-05, | |
| "loss": 1.7662, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 2.788688138256088, | |
| "eval_loss": 1.7384474277496338, | |
| "eval_runtime": 29.4245, | |
| "eval_samples_per_second": 3.399, | |
| "eval_steps_per_second": 0.442, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 2.8279654359780046, | |
| "grad_norm": 0.23036067187786102, | |
| "learning_rate": 3.1859179517020655e-05, | |
| "loss": 1.7557, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.8279654359780046, | |
| "eval_loss": 1.7396867275238037, | |
| "eval_runtime": 29.4237, | |
| "eval_samples_per_second": 3.399, | |
| "eval_steps_per_second": 0.442, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.8672427336999213, | |
| "grad_norm": 0.2239353060722351, | |
| "learning_rate": 2.458539423916206e-05, | |
| "loss": 1.7557, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 2.8672427336999213, | |
| "eval_loss": 1.7404838800430298, | |
| "eval_runtime": 29.3962, | |
| "eval_samples_per_second": 3.402, | |
| "eval_steps_per_second": 0.442, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 2.906520031421838, | |
| "grad_norm": 0.2530839145183563, | |
| "learning_rate": 1.7311608961303462e-05, | |
| "loss": 1.753, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.906520031421838, | |
| "eval_loss": 1.7404457330703735, | |
| "eval_runtime": 29.4897, | |
| "eval_samples_per_second": 3.391, | |
| "eval_steps_per_second": 0.441, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.9457973291437547, | |
| "grad_norm": 0.2325269728899002, | |
| "learning_rate": 1.0037823683444865e-05, | |
| "loss": 1.7788, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 2.9457973291437547, | |
| "eval_loss": 1.7380776405334473, | |
| "eval_runtime": 29.4462, | |
| "eval_samples_per_second": 3.396, | |
| "eval_steps_per_second": 0.441, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 2.9850746268656714, | |
| "grad_norm": 0.23328797519207, | |
| "learning_rate": 2.764038405586267e-06, | |
| "loss": 1.7539, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.9850746268656714, | |
| "eval_loss": 1.73746657371521, | |
| "eval_runtime": 29.4954, | |
| "eval_samples_per_second": 3.39, | |
| "eval_steps_per_second": 0.441, | |
| "step": 3800 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 3819, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.78476511570944e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |