| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.995276334435522, | |
| "eval_steps": 50, | |
| "global_step": 352, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "dpo_loss": 0.6931471824645996, | |
| "epoch": 0.005668398677373642, | |
| "grad_norm": 1341.3687944189992, | |
| "learning_rate": 1.3888888888888888e-07, | |
| "logits": -1.3147305250167847, | |
| "logps": -88.0877456665039, | |
| "loss": 0.6931, | |
| "objective": 0.6931471824645996, | |
| "ranking_idealized": 0.5208333134651184, | |
| "ranking_idealized_expo": 0.5208333134651184, | |
| "ranking_simple": 0.5208333134651184, | |
| "regularize": 0.6931471824645996, | |
| "step": 1 | |
| }, | |
| { | |
| "dpo_loss": 1.0099722146987915, | |
| "epoch": 0.02834199338686821, | |
| "grad_norm": 2182.754930022205, | |
| "learning_rate": 6.944444444444446e-07, | |
| "logits": -1.367998719215393, | |
| "logps": -84.27975463867188, | |
| "loss": 0.9543, | |
| "objective": 1.0099722146987915, | |
| "ranking_idealized": 0.546875, | |
| "ranking_idealized_expo": 0.546875, | |
| "ranking_simple": 0.546875, | |
| "regularize": 1.0099722146987915, | |
| "step": 5 | |
| }, | |
| { | |
| "dpo_loss": 1.232261061668396, | |
| "epoch": 0.05668398677373642, | |
| "grad_norm": 1955.6998628854572, | |
| "learning_rate": 1.3888888888888892e-06, | |
| "logits": -1.4522252082824707, | |
| "logps": -83.40966796875, | |
| "loss": 1.4248, | |
| "objective": 1.232261061668396, | |
| "ranking_idealized": 0.512499988079071, | |
| "ranking_idealized_expo": 0.5083333253860474, | |
| "ranking_simple": 0.512499988079071, | |
| "regularize": 1.232261061668396, | |
| "step": 10 | |
| }, | |
| { | |
| "dpo_loss": 3.4686033725738525, | |
| "epoch": 0.08502598016060463, | |
| "grad_norm": 2014.579294209837, | |
| "learning_rate": 2.0833333333333334e-06, | |
| "logits": -1.4239295721054077, | |
| "logps": -83.23849487304688, | |
| "loss": 2.5533, | |
| "objective": 3.4686033725738525, | |
| "ranking_idealized": 0.5041666626930237, | |
| "ranking_idealized_expo": 0.5041666626930237, | |
| "ranking_simple": 0.5083333253860474, | |
| "regularize": 3.4686033725738525, | |
| "step": 15 | |
| }, | |
| { | |
| "dpo_loss": 4.306813716888428, | |
| "epoch": 0.11336797354747284, | |
| "grad_norm": 1483.6608319475595, | |
| "learning_rate": 2.7777777777777783e-06, | |
| "logits": -1.3964630365371704, | |
| "logps": -84.55473327636719, | |
| "loss": 3.9251, | |
| "objective": 4.306813716888428, | |
| "ranking_idealized": 0.5375000238418579, | |
| "ranking_idealized_expo": 0.5375000238418579, | |
| "ranking_simple": 0.550000011920929, | |
| "regularize": 4.306813716888428, | |
| "step": 20 | |
| }, | |
| { | |
| "dpo_loss": 4.465227127075195, | |
| "epoch": 0.14170996693434104, | |
| "grad_norm": 1726.5319247739817, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "logits": -1.406508207321167, | |
| "logps": -83.8473129272461, | |
| "loss": 5.3642, | |
| "objective": 4.465227127075195, | |
| "ranking_idealized": 0.4958333373069763, | |
| "ranking_idealized_expo": 0.4958333373069763, | |
| "ranking_simple": 0.5083333253860474, | |
| "regularize": 4.465227127075195, | |
| "step": 25 | |
| }, | |
| { | |
| "dpo_loss": 7.850214004516602, | |
| "epoch": 0.17005196032120926, | |
| "grad_norm": 1356.804024894421, | |
| "learning_rate": 4.166666666666667e-06, | |
| "logits": -1.386705756187439, | |
| "logps": -84.02658081054688, | |
| "loss": 6.5533, | |
| "objective": 7.850214004516602, | |
| "ranking_idealized": 0.5416666865348816, | |
| "ranking_idealized_expo": 0.5416666865348816, | |
| "ranking_simple": 0.550000011920929, | |
| "regularize": 7.850214004516602, | |
| "step": 30 | |
| }, | |
| { | |
| "dpo_loss": 7.777320384979248, | |
| "epoch": 0.19839395370807747, | |
| "grad_norm": 1260.3392527439414, | |
| "learning_rate": 4.861111111111111e-06, | |
| "logits": -1.3726133108139038, | |
| "logps": -81.83206176757812, | |
| "loss": 6.6118, | |
| "objective": 7.777320384979248, | |
| "ranking_idealized": 0.5041666626930237, | |
| "ranking_idealized_expo": 0.5041666626930237, | |
| "ranking_simple": 0.5291666388511658, | |
| "regularize": 7.777320384979248, | |
| "step": 35 | |
| }, | |
| { | |
| "dpo_loss": 6.831168174743652, | |
| "epoch": 0.22673594709494568, | |
| "grad_norm": 1252.0799000439422, | |
| "learning_rate": 4.998023493068255e-06, | |
| "logits": -1.359412431716919, | |
| "logps": -81.25328826904297, | |
| "loss": 7.7682, | |
| "objective": 6.831168174743652, | |
| "ranking_idealized": 0.5208333134651184, | |
| "ranking_idealized_expo": 0.5208333134651184, | |
| "ranking_simple": 0.5625, | |
| "regularize": 6.831168174743652, | |
| "step": 40 | |
| }, | |
| { | |
| "dpo_loss": 8.811128616333008, | |
| "epoch": 0.25507794048181387, | |
| "grad_norm": 1307.005375082768, | |
| "learning_rate": 4.989999289644993e-06, | |
| "logits": -1.483437418937683, | |
| "logps": -85.30819702148438, | |
| "loss": 8.3899, | |
| "objective": 8.811128616333008, | |
| "ranking_idealized": 0.5375000238418579, | |
| "ranking_idealized_expo": 0.5375000238418579, | |
| "ranking_simple": 0.5666666626930237, | |
| "regularize": 8.811128616333008, | |
| "step": 45 | |
| }, | |
| { | |
| "dpo_loss": 8.527603149414062, | |
| "epoch": 0.2834199338686821, | |
| "grad_norm": 1144.3338187353374, | |
| "learning_rate": 4.975823666181256e-06, | |
| "logits": -1.4239095449447632, | |
| "logps": -83.47504425048828, | |
| "loss": 9.5723, | |
| "objective": 8.527603149414062, | |
| "ranking_idealized": 0.42500001192092896, | |
| "ranking_idealized_expo": 0.42500001192092896, | |
| "ranking_simple": 0.5041666626930237, | |
| "regularize": 8.527603149414062, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2834199338686821, | |
| "eval_dpo_loss": 9.650106430053711, | |
| "eval_logits": -1.4979068040847778, | |
| "eval_logps": -89.68624114990234, | |
| "eval_loss": 9.258649826049805, | |
| "eval_objective": 9.650106430053711, | |
| "eval_ranking_idealized": 0.5092975497245789, | |
| "eval_ranking_idealized_expo": 0.5092975497245789, | |
| "eval_ranking_simple": 0.5134297609329224, | |
| "eval_regularize": 9.650106430053711, | |
| "eval_runtime": 260.9367, | |
| "eval_samples_per_second": 22.189, | |
| "eval_steps_per_second": 0.927, | |
| "step": 50 | |
| }, | |
| { | |
| "dpo_loss": 13.282089233398438, | |
| "epoch": 0.3117619272555503, | |
| "grad_norm": 1044.8496740930686, | |
| "learning_rate": 4.955531642853404e-06, | |
| "logits": -1.527586579322815, | |
| "logps": -82.80742645263672, | |
| "loss": 10.597, | |
| "objective": 13.282089233398438, | |
| "ranking_idealized": 0.5416666865348816, | |
| "ranking_idealized_expo": 0.5416666865348816, | |
| "ranking_simple": 0.5625, | |
| "regularize": 13.282089233398438, | |
| "step": 55 | |
| }, | |
| { | |
| "dpo_loss": 10.371745109558105, | |
| "epoch": 0.3401039206424185, | |
| "grad_norm": 1331.4227369121395, | |
| "learning_rate": 4.929173350101025e-06, | |
| "logits": -1.4083276987075806, | |
| "logps": -79.24844360351562, | |
| "loss": 10.3404, | |
| "objective": 10.371745109558105, | |
| "ranking_idealized": 0.5166666507720947, | |
| "ranking_idealized_expo": 0.512499988079071, | |
| "ranking_simple": 0.5625, | |
| "regularize": 10.371745109558105, | |
| "step": 60 | |
| }, | |
| { | |
| "dpo_loss": 10.49330997467041, | |
| "epoch": 0.3684459140292867, | |
| "grad_norm": 1137.902525031776, | |
| "learning_rate": 4.896813904782162e-06, | |
| "logits": -1.2618812322616577, | |
| "logps": -80.0319595336914, | |
| "loss": 11.0189, | |
| "objective": 10.49330997467041, | |
| "ranking_idealized": 0.5083333253860474, | |
| "ranking_idealized_expo": 0.5083333253860474, | |
| "ranking_simple": 0.5541666746139526, | |
| "regularize": 10.49330997467041, | |
| "step": 65 | |
| }, | |
| { | |
| "dpo_loss": 9.181135177612305, | |
| "epoch": 0.39678790741615494, | |
| "grad_norm": 1088.944949936674, | |
| "learning_rate": 4.858533249305337e-06, | |
| "logits": -1.2879325151443481, | |
| "logps": -80.44699096679688, | |
| "loss": 11.3087, | |
| "objective": 9.181135177612305, | |
| "ranking_idealized": 0.5166666507720947, | |
| "ranking_idealized_expo": 0.5166666507720947, | |
| "ranking_simple": 0.5416666865348816, | |
| "regularize": 9.181135177612305, | |
| "step": 70 | |
| }, | |
| { | |
| "dpo_loss": 10.468655586242676, | |
| "epoch": 0.42512990080302315, | |
| "grad_norm": 999.1732460095175, | |
| "learning_rate": 4.814425954135786e-06, | |
| "logits": -1.2509528398513794, | |
| "logps": -79.77677154541016, | |
| "loss": 11.3742, | |
| "objective": 10.468655586242676, | |
| "ranking_idealized": 0.5458333492279053, | |
| "ranking_idealized_expo": 0.5458333492279053, | |
| "ranking_simple": 0.5791666507720947, | |
| "regularize": 10.468655586242676, | |
| "step": 75 | |
| }, | |
| { | |
| "dpo_loss": 12.168583869934082, | |
| "epoch": 0.45347189418989137, | |
| "grad_norm": 1008.7380345506941, | |
| "learning_rate": 4.764600984163809e-06, | |
| "logits": -1.282080888748169, | |
| "logps": -77.80094909667969, | |
| "loss": 11.9338, | |
| "objective": 12.168583869934082, | |
| "ranking_idealized": 0.5166666507720947, | |
| "ranking_idealized_expo": 0.5166666507720947, | |
| "ranking_simple": 0.6000000238418579, | |
| "regularize": 12.168583869934082, | |
| "step": 80 | |
| }, | |
| { | |
| "dpo_loss": 8.932952880859375, | |
| "epoch": 0.4818138875767596, | |
| "grad_norm": 1016.915697188663, | |
| "learning_rate": 4.709181429512391e-06, | |
| "logits": -1.2718119621276855, | |
| "logps": -74.034912109375, | |
| "loss": 11.049, | |
| "objective": 8.932952880859375, | |
| "ranking_idealized": 0.5249999761581421, | |
| "ranking_idealized_expo": 0.5249999761581421, | |
| "ranking_simple": 0.5791666507720947, | |
| "regularize": 8.932952880859375, | |
| "step": 85 | |
| }, | |
| { | |
| "dpo_loss": 13.262398719787598, | |
| "epoch": 0.5101558809636277, | |
| "grad_norm": 943.5896837416125, | |
| "learning_rate": 4.648304201449153e-06, | |
| "logits": -1.1908345222473145, | |
| "logps": -76.21371459960938, | |
| "loss": 11.6717, | |
| "objective": 13.262398719787598, | |
| "ranking_idealized": 0.4791666567325592, | |
| "ranking_idealized_expo": 0.4791666567325592, | |
| "ranking_simple": 0.49166667461395264, | |
| "regularize": 13.262398719787598, | |
| "step": 90 | |
| }, | |
| { | |
| "dpo_loss": 8.800294876098633, | |
| "epoch": 0.538497874350496, | |
| "grad_norm": 1052.3099264342789, | |
| "learning_rate": 4.582119694153834e-06, | |
| "logits": -1.3108305931091309, | |
| "logps": -74.5205078125, | |
| "loss": 9.5719, | |
| "objective": 8.800294876098633, | |
| "ranking_idealized": 0.5416666865348816, | |
| "ranking_idealized_expo": 0.5416666865348816, | |
| "ranking_simple": 0.612500011920929, | |
| "regularize": 8.800294876098633, | |
| "step": 95 | |
| }, | |
| { | |
| "dpo_loss": 7.865393161773682, | |
| "epoch": 0.5668398677373642, | |
| "grad_norm": 888.472953862153, | |
| "learning_rate": 4.510791413176912e-06, | |
| "logits": -1.3763591051101685, | |
| "logps": -75.36319732666016, | |
| "loss": 9.8364, | |
| "objective": 7.865393161773682, | |
| "ranking_idealized": 0.5416666865348816, | |
| "ranking_idealized_expo": 0.5416666865348816, | |
| "ranking_simple": 0.5958333611488342, | |
| "regularize": 7.865393161773682, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5668398677373642, | |
| "eval_dpo_loss": 15.540860176086426, | |
| "eval_logits": -1.3475213050842285, | |
| "eval_logps": -79.4201431274414, | |
| "eval_loss": 15.545284271240234, | |
| "eval_objective": 15.540860176086426, | |
| "eval_ranking_idealized": 0.5092975497245789, | |
| "eval_ranking_idealized_expo": 0.5092975497245789, | |
| "eval_ranking_simple": 0.5175619721412659, | |
| "eval_regularize": 15.540860176086426, | |
| "eval_runtime": 260.1572, | |
| "eval_samples_per_second": 22.256, | |
| "eval_steps_per_second": 0.93, | |
| "step": 100 | |
| }, | |
| { | |
| "dpo_loss": 9.041933059692383, | |
| "epoch": 0.5951818611242324, | |
| "grad_norm": 887.351501300793, | |
| "learning_rate": 4.4344955715072344e-06, | |
| "logits": -1.1800312995910645, | |
| "logps": -76.02098083496094, | |
| "loss": 10.6979, | |
| "objective": 9.041933059692383, | |
| "ranking_idealized": 0.5375000238418579, | |
| "ranking_idealized_expo": 0.5333333611488342, | |
| "ranking_simple": 0.6083333492279053, | |
| "regularize": 9.041933059692383, | |
| "step": 105 | |
| }, | |
| { | |
| "dpo_loss": 10.477563858032227, | |
| "epoch": 0.6235238545111006, | |
| "grad_norm": 852.2946697576, | |
| "learning_rate": 4.353420654246546e-06, | |
| "logits": -1.2490488290786743, | |
| "logps": -74.80319213867188, | |
| "loss": 9.0671, | |
| "objective": 10.477563858032227, | |
| "ranking_idealized": 0.44999998807907104, | |
| "ranking_idealized_expo": 0.44999998807907104, | |
| "ranking_simple": 0.5375000238418579, | |
| "regularize": 10.477563858032227, | |
| "step": 110 | |
| }, | |
| { | |
| "dpo_loss": 9.403600692749023, | |
| "epoch": 0.6518658478979689, | |
| "grad_norm": 841.5806761052793, | |
| "learning_rate": 4.267766952966369e-06, | |
| "logits": -1.2325801849365234, | |
| "logps": -73.80712127685547, | |
| "loss": 9.6957, | |
| "objective": 9.403600692749023, | |
| "ranking_idealized": 0.4833333194255829, | |
| "ranking_idealized_expo": 0.4833333194255829, | |
| "ranking_simple": 0.5625, | |
| "regularize": 9.403600692749023, | |
| "step": 115 | |
| }, | |
| { | |
| "dpo_loss": 12.634725570678711, | |
| "epoch": 0.680207841284837, | |
| "grad_norm": 843.1072296477777, | |
| "learning_rate": 4.177746070897593e-06, | |
| "logits": -1.159250259399414, | |
| "logps": -74.65933990478516, | |
| "loss": 10.2588, | |
| "objective": 12.634725570678711, | |
| "ranking_idealized": 0.5333333611488342, | |
| "ranking_idealized_expo": 0.5333333611488342, | |
| "ranking_simple": 0.5958333611488342, | |
| "regularize": 12.634725570678711, | |
| "step": 120 | |
| }, | |
| { | |
| "dpo_loss": 13.12292766571045, | |
| "epoch": 0.7085498346717053, | |
| "grad_norm": 839.6378874859962, | |
| "learning_rate": 4.083580400175153e-06, | |
| "logits": -1.1041706800460815, | |
| "logps": -73.04627990722656, | |
| "loss": 9.9023, | |
| "objective": 13.12292766571045, | |
| "ranking_idealized": 0.550000011920929, | |
| "ranking_idealized_expo": 0.550000011920929, | |
| "ranking_simple": 0.6041666865348816, | |
| "regularize": 13.12292766571045, | |
| "step": 125 | |
| }, | |
| { | |
| "dpo_loss": 9.77008056640625, | |
| "epoch": 0.7368918280585735, | |
| "grad_norm": 932.0089885121237, | |
| "learning_rate": 3.985502572429277e-06, | |
| "logits": -1.1575837135314941, | |
| "logps": -72.78858184814453, | |
| "loss": 9.9485, | |
| "objective": 9.77008056640625, | |
| "ranking_idealized": 0.44999998807907104, | |
| "ranking_idealized_expo": 0.44999998807907104, | |
| "ranking_simple": 0.5249999761581421, | |
| "regularize": 9.77008056640625, | |
| "step": 130 | |
| }, | |
| { | |
| "dpo_loss": 9.381123542785645, | |
| "epoch": 0.7652338214454416, | |
| "grad_norm": 845.3231606025312, | |
| "learning_rate": 3.8837548840805395e-06, | |
| "logits": -1.137967824935913, | |
| "logps": -75.26256561279297, | |
| "loss": 9.4092, | |
| "objective": 9.381123542785645, | |
| "ranking_idealized": 0.5291666388511658, | |
| "ranking_idealized_expo": 0.5291666388511658, | |
| "ranking_simple": 0.5791666507720947, | |
| "regularize": 9.381123542785645, | |
| "step": 135 | |
| }, | |
| { | |
| "dpo_loss": 8.004253387451172, | |
| "epoch": 0.7935758148323099, | |
| "grad_norm": 733.7736305772362, | |
| "learning_rate": 3.7785886977585562e-06, | |
| "logits": -1.0919610261917114, | |
| "logps": -74.39104461669922, | |
| "loss": 8.376, | |
| "objective": 8.004253387451172, | |
| "ranking_idealized": 0.5041666626930237, | |
| "ranking_idealized_expo": 0.5041666626930237, | |
| "ranking_simple": 0.5625, | |
| "regularize": 8.004253387451172, | |
| "step": 140 | |
| }, | |
| { | |
| "dpo_loss": 6.367016315460205, | |
| "epoch": 0.821917808219178, | |
| "grad_norm": 796.4959704194974, | |
| "learning_rate": 3.6702638213230344e-06, | |
| "logits": -1.115335464477539, | |
| "logps": -75.12439727783203, | |
| "loss": 7.6587, | |
| "objective": 6.367016315460205, | |
| "ranking_idealized": 0.4833333194255829, | |
| "ranking_idealized_expo": 0.4833333194255829, | |
| "ranking_simple": 0.5541666746139526, | |
| "regularize": 6.367016315460205, | |
| "step": 145 | |
| }, | |
| { | |
| "dpo_loss": 8.81261920928955, | |
| "epoch": 0.8502598016060463, | |
| "grad_norm": 773.8278413807583, | |
| "learning_rate": 3.5590478660213214e-06, | |
| "logits": -1.0867894887924194, | |
| "logps": -74.4196548461914, | |
| "loss": 8.8451, | |
| "objective": 8.81261920928955, | |
| "ranking_idealized": 0.5416666865348816, | |
| "ranking_idealized_expo": 0.5375000238418579, | |
| "ranking_simple": 0.5791666507720947, | |
| "regularize": 8.81261920928955, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8502598016060463, | |
| "eval_dpo_loss": 16.562625885009766, | |
| "eval_logits": -1.1121788024902344, | |
| "eval_logps": -82.14590454101562, | |
| "eval_loss": 16.66257095336914, | |
| "eval_objective": 16.562625885009766, | |
| "eval_ranking_idealized": 0.5092975497245789, | |
| "eval_ranking_idealized_expo": 0.5092975497245789, | |
| "eval_ranking_simple": 0.5144628286361694, | |
| "eval_regularize": 16.562625885009766, | |
| "eval_runtime": 259.9617, | |
| "eval_samples_per_second": 22.273, | |
| "eval_steps_per_second": 0.931, | |
| "step": 150 | |
| }, | |
| { | |
| "dpo_loss": 7.545068264007568, | |
| "epoch": 0.8786017949929145, | |
| "grad_norm": 757.3947735379465, | |
| "learning_rate": 3.4452155853680457e-06, | |
| "logits": -1.0470577478408813, | |
| "logps": -75.96894836425781, | |
| "loss": 7.0262, | |
| "objective": 7.545068264007568, | |
| "ranking_idealized": 0.512499988079071, | |
| "ranking_idealized_expo": 0.512499988079071, | |
| "ranking_simple": 0.6208333373069763, | |
| "regularize": 7.545068264007568, | |
| "step": 155 | |
| }, | |
| { | |
| "dpo_loss": 6.935917854309082, | |
| "epoch": 0.9069437883797827, | |
| "grad_norm": 751.9625745316074, | |
| "learning_rate": 3.32904819638017e-06, | |
| "logits": -0.919123113155365, | |
| "logps": -78.4984359741211, | |
| "loss": 7.8293, | |
| "objective": 6.935917854309082, | |
| "ranking_idealized": 0.4958333373069763, | |
| "ranking_idealized_expo": 0.4958333373069763, | |
| "ranking_simple": 0.5916666388511658, | |
| "regularize": 6.935917854309082, | |
| "step": 160 | |
| }, | |
| { | |
| "dpo_loss": 7.154492378234863, | |
| "epoch": 0.9352857817666509, | |
| "grad_norm": 658.8620194691712, | |
| "learning_rate": 3.2108326848442507e-06, | |
| "logits": -0.9658632278442383, | |
| "logps": -79.98240661621094, | |
| "loss": 7.7063, | |
| "objective": 7.154492378234863, | |
| "ranking_idealized": 0.5208333134651184, | |
| "ranking_idealized_expo": 0.5208333134651184, | |
| "ranking_simple": 0.5958333611488342, | |
| "regularize": 7.154492378234863, | |
| "step": 165 | |
| }, | |
| { | |
| "dpo_loss": 7.305918216705322, | |
| "epoch": 0.9636277751535192, | |
| "grad_norm": 788.1879535129325, | |
| "learning_rate": 3.090861096332263e-06, | |
| "logits": -1.0152103900909424, | |
| "logps": -79.22311401367188, | |
| "loss": 7.4405, | |
| "objective": 7.305918216705322, | |
| "ranking_idealized": 0.512499988079071, | |
| "ranking_idealized_expo": 0.5083333253860474, | |
| "ranking_simple": 0.5708333253860474, | |
| "regularize": 7.305918216705322, | |
| "step": 170 | |
| }, | |
| { | |
| "dpo_loss": 7.161056995391846, | |
| "epoch": 0.9919697685403873, | |
| "grad_norm": 703.4556255689452, | |
| "learning_rate": 2.9694298147174566e-06, | |
| "logits": -1.0000672340393066, | |
| "logps": -78.11359405517578, | |
| "loss": 6.5526, | |
| "objective": 7.161056995391846, | |
| "ranking_idealized": 0.574999988079071, | |
| "ranking_idealized_expo": 0.574999988079071, | |
| "ranking_simple": 0.6791666746139526, | |
| "regularize": 7.161056995391846, | |
| "step": 175 | |
| }, | |
| { | |
| "dpo_loss": 5.952729225158691, | |
| "epoch": 1.0203117619272555, | |
| "grad_norm": 665.4146360676681, | |
| "learning_rate": 2.8468388299726714e-06, | |
| "logits": -1.1097006797790527, | |
| "logps": -78.40138244628906, | |
| "loss": 4.6537, | |
| "objective": 5.952729225158691, | |
| "ranking_idealized": 0.5416666865348816, | |
| "ranking_idealized_expo": 0.5416666865348816, | |
| "ranking_simple": 0.637499988079071, | |
| "regularize": 5.952729225158691, | |
| "step": 180 | |
| }, | |
| { | |
| "dpo_loss": 4.262913703918457, | |
| "epoch": 1.0486537553141237, | |
| "grad_norm": 584.0991136785685, | |
| "learning_rate": 2.723390997059943e-06, | |
| "logits": -1.034575343132019, | |
| "logps": -76.80313110351562, | |
| "loss": 3.7788, | |
| "objective": 4.262913703918457, | |
| "ranking_idealized": 0.4833333194255829, | |
| "ranking_idealized_expo": 0.4833333194255829, | |
| "ranking_simple": 0.5791666507720947, | |
| "regularize": 4.262913703918457, | |
| "step": 185 | |
| }, | |
| { | |
| "dpo_loss": 3.1785500049591064, | |
| "epoch": 1.076995748700992, | |
| "grad_norm": 544.0621151708413, | |
| "learning_rate": 2.599391287742315e-06, | |
| "logits": -1.0769795179367065, | |
| "logps": -77.34581756591797, | |
| "loss": 3.5653, | |
| "objective": 3.1785500049591064, | |
| "ranking_idealized": 0.5541666746139526, | |
| "ranking_idealized_expo": 0.550000011920929, | |
| "ranking_simple": 0.6458333134651184, | |
| "regularize": 3.1785500049591064, | |
| "step": 190 | |
| }, | |
| { | |
| "dpo_loss": 2.1757826805114746, | |
| "epoch": 1.10533774208786, | |
| "grad_norm": 485.2987165965692, | |
| "learning_rate": 2.4751460371661763e-06, | |
| "logits": -1.0281838178634644, | |
| "logps": -76.58784484863281, | |
| "loss": 3.4138, | |
| "objective": 2.1757826805114746, | |
| "ranking_idealized": 0.5666666626930237, | |
| "ranking_idealized_expo": 0.5666666626930237, | |
| "ranking_simple": 0.6666666865348816, | |
| "regularize": 2.1757826805114746, | |
| "step": 195 | |
| }, | |
| { | |
| "dpo_loss": 2.5434367656707764, | |
| "epoch": 1.1336797354747283, | |
| "grad_norm": 579.2034510566444, | |
| "learning_rate": 2.3509621870754505e-06, | |
| "logits": -1.0624656677246094, | |
| "logps": -76.0323715209961, | |
| "loss": 3.8083, | |
| "objective": 2.5434367656707764, | |
| "ranking_idealized": 0.5083333253860474, | |
| "ranking_idealized_expo": 0.5083333253860474, | |
| "ranking_simple": 0.6208333373069763, | |
| "regularize": 2.5434367656707764, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.1336797354747283, | |
| "eval_dpo_loss": 16.32398223876953, | |
| "eval_logits": -1.0874205827713013, | |
| "eval_logps": -81.67513275146484, | |
| "eval_loss": 16.051929473876953, | |
| "eval_objective": 16.32398223876953, | |
| "eval_ranking_idealized": 0.5092975497245789, | |
| "eval_ranking_idealized_expo": 0.5092975497245789, | |
| "eval_ranking_simple": 0.5185950398445129, | |
| "eval_regularize": 16.32398223876953, | |
| "eval_runtime": 260.3079, | |
| "eval_samples_per_second": 22.243, | |
| "eval_steps_per_second": 0.93, | |
| "step": 200 | |
| }, | |
| { | |
| "dpo_loss": 4.850390434265137, | |
| "epoch": 1.1620217288615966, | |
| "grad_norm": 492.7125988147346, | |
| "learning_rate": 2.2271465275271985e-06, | |
| "logits": -1.0401540994644165, | |
| "logps": -76.98418426513672, | |
| "loss": 3.4645, | |
| "objective": 4.850390434265137, | |
| "ranking_idealized": 0.48750001192092896, | |
| "ranking_idealized_expo": 0.48750001192092896, | |
| "ranking_simple": 0.6041666865348816, | |
| "regularize": 4.850390434265137, | |
| "step": 205 | |
| }, | |
| { | |
| "dpo_loss": 3.4442708492279053, | |
| "epoch": 1.1903637222484649, | |
| "grad_norm": 560.4210922655933, | |
| "learning_rate": 2.1040049389819628e-06, | |
| "logits": -1.0271131992340088, | |
| "logps": -76.39945220947266, | |
| "loss": 3.425, | |
| "objective": 3.4442708492279053, | |
| "ranking_idealized": 0.5, | |
| "ranking_idealized_expo": 0.4958333373069763, | |
| "ranking_simple": 0.5874999761581421, | |
| "regularize": 3.4442708492279053, | |
| "step": 210 | |
| }, | |
| { | |
| "dpo_loss": 4.348363399505615, | |
| "epoch": 1.2187057156353331, | |
| "grad_norm": 570.4040655884006, | |
| "learning_rate": 1.9818416366412277e-06, | |
| "logits": -0.9843409657478333, | |
| "logps": -75.80220031738281, | |
| "loss": 3.8326, | |
| "objective": 4.348363399505615, | |
| "ranking_idealized": 0.4833333194255829, | |
| "ranking_idealized_expo": 0.4833333194255829, | |
| "ranking_simple": 0.612500011920929, | |
| "regularize": 4.348363399505615, | |
| "step": 215 | |
| }, | |
| { | |
| "dpo_loss": 3.9427292346954346, | |
| "epoch": 1.2470477090222012, | |
| "grad_norm": 674.1094626776156, | |
| "learning_rate": 1.8609584188988135e-06, | |
| "logits": -0.8341898322105408, | |
| "logps": -75.6758804321289, | |
| "loss": 3.8234, | |
| "objective": 3.9427292346954346, | |
| "ranking_idealized": 0.6000000238418579, | |
| "ranking_idealized_expo": 0.6000000238418579, | |
| "ranking_simple": 0.675000011920929, | |
| "regularize": 3.9427292346954346, | |
| "step": 220 | |
| }, | |
| { | |
| "dpo_loss": 3.127484083175659, | |
| "epoch": 1.2753897024090695, | |
| "grad_norm": 586.9637333254279, | |
| "learning_rate": 1.7416539217628792e-06, | |
| "logits": -0.8649470806121826, | |
| "logps": -77.80410766601562, | |
| "loss": 3.7703, | |
| "objective": 3.127484083175659, | |
| "ranking_idealized": 0.5291666388511658, | |
| "ranking_idealized_expo": 0.5291666388511658, | |
| "ranking_simple": 0.6541666388511658, | |
| "regularize": 3.127484083175659, | |
| "step": 225 | |
| }, | |
| { | |
| "dpo_loss": 3.9277312755584717, | |
| "epoch": 1.3037316957959377, | |
| "grad_norm": 500.9237074632472, | |
| "learning_rate": 1.6242228810904393e-06, | |
| "logits": -0.806861162185669, | |
| "logps": -77.2365951538086, | |
| "loss": 3.2759, | |
| "objective": 3.9277312755584717, | |
| "ranking_idealized": 0.5083333253860474, | |
| "ranking_idealized_expo": 0.5083333253860474, | |
| "ranking_simple": 0.5833333134651184, | |
| "regularize": 3.9277312755584717, | |
| "step": 230 | |
| }, | |
| { | |
| "dpo_loss": 2.1227903366088867, | |
| "epoch": 1.3320736891828058, | |
| "grad_norm": 529.1823240184154, | |
| "learning_rate": 1.508955404457015e-06, | |
| "logits": -0.8957360982894897, | |
| "logps": -74.15618133544922, | |
| "loss": 3.8217, | |
| "objective": 2.1227903366088867, | |
| "ranking_idealized": 0.4833333194255829, | |
| "ranking_idealized_expo": 0.4833333194255829, | |
| "ranking_simple": 0.6416666507720947, | |
| "regularize": 2.1227903366088867, | |
| "step": 235 | |
| }, | |
| { | |
| "dpo_loss": 3.4101717472076416, | |
| "epoch": 1.360415682569674, | |
| "grad_norm": 527.5490363089041, | |
| "learning_rate": 1.3961362544602215e-06, | |
| "logits": -0.8434502482414246, | |
| "logps": -76.20458221435547, | |
| "loss": 3.1011, | |
| "objective": 3.4101717472076416, | |
| "ranking_idealized": 0.4791666567325592, | |
| "ranking_idealized_expo": 0.4791666567325592, | |
| "ranking_simple": 0.625, | |
| "regularize": 3.4101717472076416, | |
| "step": 240 | |
| }, | |
| { | |
| "dpo_loss": 3.476386070251465, | |
| "epoch": 1.3887576759565423, | |
| "grad_norm": 571.2307522066004, | |
| "learning_rate": 1.2860441452278578e-06, | |
| "logits": -0.8681662082672119, | |
| "logps": -76.64582061767578, | |
| "loss": 3.8625, | |
| "objective": 3.476386070251465, | |
| "ranking_idealized": 0.5208333134651184, | |
| "ranking_idealized_expo": 0.5166666507720947, | |
| "ranking_simple": 0.612500011920929, | |
| "regularize": 3.476386070251465, | |
| "step": 245 | |
| }, | |
| { | |
| "dpo_loss": 1.8152854442596436, | |
| "epoch": 1.4170996693434104, | |
| "grad_norm": 506.7476965423747, | |
| "learning_rate": 1.1789510538684524e-06, | |
| "logits": -0.8697729110717773, | |
| "logps": -76.35972595214844, | |
| "loss": 3.6019, | |
| "objective": 1.8152854442596436, | |
| "ranking_idealized": 0.47083333134651184, | |
| "ranking_idealized_expo": 0.47083333134651184, | |
| "ranking_simple": 0.612500011920929, | |
| "regularize": 1.8152854442596436, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.4170996693434104, | |
| "eval_dpo_loss": 15.767914772033691, | |
| "eval_logits": -0.993295431137085, | |
| "eval_logps": -81.56088256835938, | |
| "eval_loss": 15.814361572265625, | |
| "eval_objective": 15.767914772033691, | |
| "eval_ranking_idealized": 0.5092975497245789, | |
| "eval_ranking_idealized_expo": 0.5092975497245789, | |
| "eval_ranking_simple": 0.5175619721412659, | |
| "eval_regularize": 15.767914772033691, | |
| "eval_runtime": 259.6498, | |
| "eval_samples_per_second": 22.299, | |
| "eval_steps_per_second": 0.932, | |
| "step": 250 | |
| }, | |
| { | |
| "dpo_loss": 3.3358986377716064, | |
| "epoch": 1.4454416627302786, | |
| "grad_norm": 563.3612521264971, | |
| "learning_rate": 1.0751215485652644e-06, | |
| "logits": -0.891727089881897, | |
| "logps": -76.76223754882812, | |
| "loss": 3.0719, | |
| "objective": 3.3358986377716064, | |
| "ranking_idealized": 0.5458333492279053, | |
| "ranking_idealized_expo": 0.5458333492279053, | |
| "ranking_simple": 0.637499988079071, | |
| "regularize": 3.3358986377716064, | |
| "step": 255 | |
| }, | |
| { | |
| "dpo_loss": 2.244086265563965, | |
| "epoch": 1.473783656117147, | |
| "grad_norm": 518.9208628864491, | |
| "learning_rate": 9.74812134973689e-07, | |
| "logits": -0.9449579119682312, | |
| "logps": -76.72704315185547, | |
| "loss": 2.8837, | |
| "objective": 2.244086265563965, | |
| "ranking_idealized": 0.48750001192092896, | |
| "ranking_idealized_expo": 0.48750001192092896, | |
| "ranking_simple": 0.6041666865348816, | |
| "regularize": 2.244086265563965, | |
| "step": 260 | |
| }, | |
| { | |
| "dpo_loss": 3.1619741916656494, | |
| "epoch": 1.5021256495040152, | |
| "grad_norm": 583.760652002654, | |
| "learning_rate": 8.78270622536716e-07, | |
| "logits": -0.953507125377655, | |
| "logps": -75.11036682128906, | |
| "loss": 2.925, | |
| "objective": 3.1619741916656494, | |
| "ranking_idealized": 0.5041666626930237, | |
| "ranking_idealized_expo": 0.5041666626930237, | |
| "ranking_simple": 0.6083333492279053, | |
| "regularize": 3.1619741916656494, | |
| "step": 265 | |
| }, | |
| { | |
| "dpo_loss": 2.98382830619812, | |
| "epoch": 1.5304676428908834, | |
| "grad_norm": 520.7932622156393, | |
| "learning_rate": 7.857355122839674e-07, | |
| "logits": -0.9570529460906982, | |
| "logps": -76.34257507324219, | |
| "loss": 2.3431, | |
| "objective": 2.98382830619812, | |
| "ranking_idealized": 0.5083333253860474, | |
| "ranking_idealized_expo": 0.5041666626930237, | |
| "ranking_simple": 0.6041666865348816, | |
| "regularize": 2.98382830619812, | |
| "step": 270 | |
| }, | |
| { | |
| "dpo_loss": 2.821234941482544, | |
| "epoch": 1.5588096362777515, | |
| "grad_norm": 536.6516299894466, | |
| "learning_rate": 6.974354076267081e-07, | |
| "logits": -0.9497679471969604, | |
| "logps": -76.84500122070312, | |
| "loss": 3.0543, | |
| "objective": 2.821234941482544, | |
| "ranking_idealized": 0.5249999761581421, | |
| "ranking_idealized_expo": 0.5249999761581421, | |
| "ranking_simple": 0.6625000238418579, | |
| "regularize": 2.821234941482544, | |
| "step": 275 | |
| }, | |
| { | |
| "dpo_loss": 1.6416581869125366, | |
| "epoch": 1.5871516296646198, | |
| "grad_norm": 473.79557869240267, | |
| "learning_rate": 6.135884496044245e-07, | |
| "logits": -0.8786503076553345, | |
| "logps": -75.35763549804688, | |
| "loss": 2.5903, | |
| "objective": 1.6416581869125366, | |
| "ranking_idealized": 0.5041666626930237, | |
| "ranking_idealized_expo": 0.5041666626930237, | |
| "ranking_simple": 0.6000000238418579, | |
| "regularize": 1.6416581869125366, | |
| "step": 280 | |
| }, | |
| { | |
| "dpo_loss": 2.1852915287017822, | |
| "epoch": 1.615493623051488, | |
| "grad_norm": 530.1117274697609, | |
| "learning_rate": 5.344017779781835e-07, | |
| "logits": -0.9772613048553467, | |
| "logps": -76.81114196777344, | |
| "loss": 2.5659, | |
| "objective": 2.1852915287017822, | |
| "ranking_idealized": 0.5166666507720947, | |
| "ranking_idealized_expo": 0.5166666507720947, | |
| "ranking_simple": 0.6166666746139526, | |
| "regularize": 2.1852915287017822, | |
| "step": 285 | |
| }, | |
| { | |
| "dpo_loss": 3.540595769882202, | |
| "epoch": 1.643835616438356, | |
| "grad_norm": 543.628954255078, | |
| "learning_rate": 4.6007101950209827e-07, | |
| "logits": -0.9982803463935852, | |
| "logps": -74.47586059570312, | |
| "loss": 3.1645, | |
| "objective": 3.540595769882202, | |
| "ranking_idealized": 0.5625, | |
| "ranking_idealized_expo": 0.5625, | |
| "ranking_simple": 0.625, | |
| "regularize": 3.540595769882202, | |
| "step": 290 | |
| }, | |
| { | |
| "dpo_loss": 1.856958031654358, | |
| "epoch": 1.6721776098252243, | |
| "grad_norm": 523.9882241509144, | |
| "learning_rate": 3.907798046371139e-07, | |
| "logits": -0.9143877029418945, | |
| "logps": -73.36082458496094, | |
| "loss": 2.4952, | |
| "objective": 1.856958031654358, | |
| "ranking_idealized": 0.4791666567325592, | |
| "ranking_idealized_expo": 0.4791666567325592, | |
| "ranking_simple": 0.612500011920929, | |
| "regularize": 1.856958031654358, | |
| "step": 295 | |
| }, | |
| { | |
| "dpo_loss": 1.6541905403137207, | |
| "epoch": 1.7005196032120926, | |
| "grad_norm": 474.76333707639907, | |
| "learning_rate": 3.266993139010438e-07, | |
| "logits": -0.9617911577224731, | |
| "logps": -74.96733093261719, | |
| "loss": 2.1682, | |
| "objective": 1.6541905403137207, | |
| "ranking_idealized": 0.574999988079071, | |
| "ranking_idealized_expo": 0.574999988079071, | |
| "ranking_simple": 0.6708333492279053, | |
| "regularize": 1.6541905403137207, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.7005196032120926, | |
| "eval_dpo_loss": 15.200382232666016, | |
| "eval_logits": -1.0035560131072998, | |
| "eval_logps": -80.33287811279297, | |
| "eval_loss": 15.382392883300781, | |
| "eval_objective": 15.200382232666016, | |
| "eval_ranking_idealized": 0.5092975497245789, | |
| "eval_ranking_idealized_expo": 0.5092975497245789, | |
| "eval_ranking_simple": 0.5113636255264282, | |
| "eval_regularize": 15.200382232666016, | |
| "eval_runtime": 264.8415, | |
| "eval_samples_per_second": 21.862, | |
| "eval_steps_per_second": 0.914, | |
| "step": 300 | |
| }, | |
| { | |
| "dpo_loss": 2.5656979084014893, | |
| "epoch": 1.7288615965989607, | |
| "grad_norm": 503.0202503507287, | |
| "learning_rate": 2.6798785497556184e-07, | |
| "logits": -0.933074414730072, | |
| "logps": -76.0098876953125, | |
| "loss": 2.728, | |
| "objective": 2.5656979084014893, | |
| "ranking_idealized": 0.5083333253860474, | |
| "ranking_idealized_expo": 0.5041666626930237, | |
| "ranking_simple": 0.6166666746139526, | |
| "regularize": 2.5656979084014893, | |
| "step": 305 | |
| }, | |
| { | |
| "dpo_loss": 2.7561254501342773, | |
| "epoch": 1.7572035899858292, | |
| "grad_norm": 551.8103525795702, | |
| "learning_rate": 2.1479047161491351e-07, | |
| "logits": -0.9786698222160339, | |
| "logps": -73.98075103759766, | |
| "loss": 3.2082, | |
| "objective": 2.7561254501342773, | |
| "ranking_idealized": 0.5208333134651184, | |
| "ranking_idealized_expo": 0.5208333134651184, | |
| "ranking_simple": 0.625, | |
| "regularize": 2.7561254501342773, | |
| "step": 310 | |
| }, | |
| { | |
| "dpo_loss": 2.8616750240325928, | |
| "epoch": 1.7855455833726972, | |
| "grad_norm": 481.44311587523885, | |
| "learning_rate": 1.6723858532249783e-07, | |
| "logits": -0.9176926016807556, | |
| "logps": -74.52711486816406, | |
| "loss": 2.6534, | |
| "objective": 2.8616750240325928, | |
| "ranking_idealized": 0.5083333253860474, | |
| "ranking_idealized_expo": 0.5083333253860474, | |
| "ranking_simple": 0.6208333373069763, | |
| "regularize": 2.8616750240325928, | |
| "step": 315 | |
| }, | |
| { | |
| "dpo_loss": 2.875875949859619, | |
| "epoch": 1.8138875767595655, | |
| "grad_norm": 473.64902583279064, | |
| "learning_rate": 1.2544967068054332e-07, | |
| "logits": -0.968319833278656, | |
| "logps": -76.5433120727539, | |
| "loss": 2.4726, | |
| "objective": 2.875875949859619, | |
| "ranking_idealized": 0.5041666626930237, | |
| "ranking_idealized_expo": 0.5041666626930237, | |
| "ranking_simple": 0.6083333492279053, | |
| "regularize": 2.875875949859619, | |
| "step": 320 | |
| }, | |
| { | |
| "dpo_loss": 2.392838716506958, | |
| "epoch": 1.8422295701464337, | |
| "grad_norm": 598.6261084207288, | |
| "learning_rate": 8.952696513496756e-08, | |
| "logits": -1.005333662033081, | |
| "logps": -77.68478393554688, | |
| "loss": 2.8906, | |
| "objective": 2.392838716506958, | |
| "ranking_idealized": 0.5166666507720947, | |
| "ranking_idealized_expo": 0.5166666507720947, | |
| "ranking_simple": 0.6541666388511658, | |
| "regularize": 2.392838716506958, | |
| "step": 325 | |
| }, | |
| { | |
| "dpo_loss": 2.538822889328003, | |
| "epoch": 1.8705715635333018, | |
| "grad_norm": 468.16090350396775, | |
| "learning_rate": 5.955921395237318e-08, | |
| "logits": -0.8062960505485535, | |
| "logps": -74.2054443359375, | |
| "loss": 2.6483, | |
| "objective": 2.538822889328003, | |
| "ranking_idealized": 0.4749999940395355, | |
| "ranking_idealized_expo": 0.47083333134651184, | |
| "ranking_simple": 0.5791666507720947, | |
| "regularize": 2.538822889328003, | |
| "step": 330 | |
| }, | |
| { | |
| "dpo_loss": 2.293182849884033, | |
| "epoch": 1.89891355692017, | |
| "grad_norm": 457.5310635009536, | |
| "learning_rate": 3.5620450979259034e-08, | |
| "logits": -0.9471557140350342, | |
| "logps": -74.38040161132812, | |
| "loss": 2.5969, | |
| "objective": 2.293182849884033, | |
| "ranking_idealized": 0.4958333373069763, | |
| "ranking_idealized_expo": 0.4958333373069763, | |
| "ranking_simple": 0.6333333253860474, | |
| "regularize": 2.293182849884033, | |
| "step": 335 | |
| }, | |
| { | |
| "dpo_loss": 2.192481279373169, | |
| "epoch": 1.9272555503070383, | |
| "grad_norm": 536.3353914926571, | |
| "learning_rate": 1.7769815745066476e-08, | |
| "logits": -0.9198692440986633, | |
| "logps": -76.65982055664062, | |
| "loss": 2.8401, | |
| "objective": 2.192481279373169, | |
| "ranking_idealized": 0.5416666865348816, | |
| "ranking_idealized_expo": 0.5416666865348816, | |
| "ranking_simple": 0.6625000238418579, | |
| "regularize": 2.192481279373169, | |
| "step": 340 | |
| }, | |
| { | |
| "dpo_loss": 1.9803303480148315, | |
| "epoch": 1.9555975436939064, | |
| "grad_norm": 445.05103819649605, | |
| "learning_rate": 6.051407360895822e-09, | |
| "logits": -0.9369055032730103, | |
| "logps": -75.275634765625, | |
| "loss": 2.7774, | |
| "objective": 1.9803303480148315, | |
| "ranking_idealized": 0.5333333611488342, | |
| "ranking_idealized_expo": 0.5333333611488342, | |
| "ranking_simple": 0.6333333253860474, | |
| "regularize": 1.9803303480148315, | |
| "step": 345 | |
| }, | |
| { | |
| "dpo_loss": 2.6546401977539062, | |
| "epoch": 1.9839395370807746, | |
| "grad_norm": 571.7789309975336, | |
| "learning_rate": 4.941755748361088e-10, | |
| "logits": -0.8779342770576477, | |
| "logps": -76.04293060302734, | |
| "loss": 2.703, | |
| "objective": 2.6546401977539062, | |
| "ranking_idealized": 0.48750001192092896, | |
| "ranking_idealized_expo": 0.48750001192092896, | |
| "ranking_simple": 0.5874999761581421, | |
| "regularize": 2.6546401977539062, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.9839395370807746, | |
| "eval_dpo_loss": 15.144511222839355, | |
| "eval_logits": -1.0045602321624756, | |
| "eval_logps": -80.39806365966797, | |
| "eval_loss": 15.256567001342773, | |
| "eval_objective": 15.144511222839355, | |
| "eval_ranking_idealized": 0.5092975497245789, | |
| "eval_ranking_idealized_expo": 0.5092975497245789, | |
| "eval_ranking_simple": 0.5134297609329224, | |
| "eval_regularize": 15.144511222839355, | |
| "eval_runtime": 259.6567, | |
| "eval_samples_per_second": 22.299, | |
| "eval_steps_per_second": 0.932, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.995276334435522, | |
| "step": 352, | |
| "total_flos": 0.0, | |
| "train_loss": 5.737799351865595, | |
| "train_runtime": 14206.9508, | |
| "train_samples_per_second": 7.152, | |
| "train_steps_per_second": 0.025 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 352, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |