| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9990186457311089, | |
| "eval_steps": 100, | |
| "global_step": 509, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001962708537782139, | |
| "grad_norm": 2.4117076017287205, | |
| "learning_rate": 9.803921568627451e-09, | |
| "logits/chosen": -1.125, | |
| "logits/rejected": -1.1875, | |
| "logps/chosen": -500.0, | |
| "logps/chosen_bottom_tokens": -14.5, | |
| "logps/chosen_top_tokens": -0.0005645751953125, | |
| "logps/rejected": -520.0, | |
| "logps/rejected_bottom_tokens": -13.9375, | |
| "logps/rejected_top_tokens": -0.00054168701171875, | |
| "loss": 0.6914, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.019627085377821395, | |
| "grad_norm": 2.3800058601187866, | |
| "learning_rate": 9.80392156862745e-08, | |
| "logits/chosen": -1.1640625, | |
| "logits/rejected": -1.203125, | |
| "logps/chosen": -380.0, | |
| "logps/chosen_bottom_tokens": -14.125, | |
| "logps/chosen_top_tokens": -0.000804901123046875, | |
| "logps/rejected": -316.0, | |
| "logps/rejected_bottom_tokens": -14.125, | |
| "logps/rejected_top_tokens": -0.000827789306640625, | |
| "loss": 0.6922, | |
| "rewards/accuracies": 0.41111111640930176, | |
| "rewards/chosen": -0.000202178955078125, | |
| "rewards/margins": 0.0035247802734375, | |
| "rewards/rejected": -0.00372314453125, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03925417075564279, | |
| "grad_norm": 2.4064882227881057, | |
| "learning_rate": 1.96078431372549e-07, | |
| "logits/chosen": -1.0859375, | |
| "logits/rejected": -1.1484375, | |
| "logps/chosen": -374.0, | |
| "logps/chosen_bottom_tokens": -14.125, | |
| "logps/chosen_top_tokens": -0.000835418701171875, | |
| "logps/rejected": -324.0, | |
| "logps/rejected_bottom_tokens": -14.0625, | |
| "logps/rejected_top_tokens": -0.00084686279296875, | |
| "loss": 0.6913, | |
| "rewards/accuracies": 0.4399999976158142, | |
| "rewards/chosen": 0.003662109375, | |
| "rewards/margins": 0.0033111572265625, | |
| "rewards/rejected": 0.0003528594970703125, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.058881256133464184, | |
| "grad_norm": 2.3536995350535426, | |
| "learning_rate": 2.941176470588235e-07, | |
| "logits/chosen": -1.125, | |
| "logits/rejected": -1.1796875, | |
| "logps/chosen": -364.0, | |
| "logps/chosen_bottom_tokens": -14.25, | |
| "logps/chosen_top_tokens": -0.000762939453125, | |
| "logps/rejected": -324.0, | |
| "logps/rejected_bottom_tokens": -14.125, | |
| "logps/rejected_top_tokens": -0.00078582763671875, | |
| "loss": 0.6938, | |
| "rewards/accuracies": 0.3850000202655792, | |
| "rewards/chosen": -0.0030670166015625, | |
| "rewards/margins": -0.0067138671875, | |
| "rewards/rejected": 0.003631591796875, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07850834151128558, | |
| "grad_norm": 2.3870217018270155, | |
| "learning_rate": 3.92156862745098e-07, | |
| "logits/chosen": -1.125, | |
| "logits/rejected": -1.15625, | |
| "logps/chosen": -378.0, | |
| "logps/chosen_bottom_tokens": -14.1875, | |
| "logps/chosen_top_tokens": -0.000759124755859375, | |
| "logps/rejected": -338.0, | |
| "logps/rejected_bottom_tokens": -14.125, | |
| "logps/rejected_top_tokens": -0.000804901123046875, | |
| "loss": 0.6929, | |
| "rewards/accuracies": 0.445000022649765, | |
| "rewards/chosen": 0.000881195068359375, | |
| "rewards/margins": 0.00244140625, | |
| "rewards/rejected": -0.00154876708984375, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09813542688910697, | |
| "grad_norm": 2.4788478916800147, | |
| "learning_rate": 4.901960784313725e-07, | |
| "logits/chosen": -1.1171875, | |
| "logits/rejected": -1.1484375, | |
| "logps/chosen": -406.0, | |
| "logps/chosen_bottom_tokens": -14.0625, | |
| "logps/chosen_top_tokens": -0.000774383544921875, | |
| "logps/rejected": -352.0, | |
| "logps/rejected_bottom_tokens": -14.0, | |
| "logps/rejected_top_tokens": -0.00080108642578125, | |
| "loss": 0.6915, | |
| "rewards/accuracies": 0.5400000214576721, | |
| "rewards/chosen": 0.0026397705078125, | |
| "rewards/margins": 0.005889892578125, | |
| "rewards/rejected": -0.00323486328125, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.11776251226692837, | |
| "grad_norm": 2.360316334548125, | |
| "learning_rate": 4.995237599803335e-07, | |
| "logits/chosen": -1.140625, | |
| "logits/rejected": -1.203125, | |
| "logps/chosen": -406.0, | |
| "logps/chosen_bottom_tokens": -14.1875, | |
| "logps/chosen_top_tokens": -0.000782012939453125, | |
| "logps/rejected": -322.0, | |
| "logps/rejected_bottom_tokens": -14.0625, | |
| "logps/rejected_top_tokens": -0.0008087158203125, | |
| "loss": 0.6913, | |
| "rewards/accuracies": 0.42500001192092896, | |
| "rewards/chosen": 0.00244140625, | |
| "rewards/margins": 0.00201416015625, | |
| "rewards/rejected": 0.000431060791015625, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.13738959764474976, | |
| "grad_norm": 2.3051434353276847, | |
| "learning_rate": 4.978798275112142e-07, | |
| "logits/chosen": -1.09375, | |
| "logits/rejected": -1.1328125, | |
| "logps/chosen": -372.0, | |
| "logps/chosen_bottom_tokens": -14.0625, | |
| "logps/chosen_top_tokens": -0.00078582763671875, | |
| "logps/rejected": -330.0, | |
| "logps/rejected_bottom_tokens": -14.0625, | |
| "logps/rejected_top_tokens": -0.000789642333984375, | |
| "loss": 0.688, | |
| "rewards/accuracies": 0.5049999952316284, | |
| "rewards/chosen": 0.00897216796875, | |
| "rewards/margins": 0.01190185546875, | |
| "rewards/rejected": -0.0028839111328125, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.15701668302257116, | |
| "grad_norm": 2.2866846976386, | |
| "learning_rate": 4.950700530747689e-07, | |
| "logits/chosen": -1.078125, | |
| "logits/rejected": -1.1484375, | |
| "logps/chosen": -378.0, | |
| "logps/chosen_bottom_tokens": -14.0, | |
| "logps/chosen_top_tokens": -0.000934600830078125, | |
| "logps/rejected": -308.0, | |
| "logps/rejected_bottom_tokens": -14.0, | |
| "logps/rejected_top_tokens": -0.00087738037109375, | |
| "loss": 0.685, | |
| "rewards/accuracies": 0.5450000166893005, | |
| "rewards/chosen": -0.00121307373046875, | |
| "rewards/margins": 0.01483154296875, | |
| "rewards/rejected": -0.01611328125, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.17664376840039253, | |
| "grad_norm": 2.3053347338418098, | |
| "learning_rate": 4.911076517558622e-07, | |
| "logits/chosen": -1.125, | |
| "logits/rejected": -1.15625, | |
| "logps/chosen": -382.0, | |
| "logps/chosen_bottom_tokens": -14.125, | |
| "logps/chosen_top_tokens": -0.000823974609375, | |
| "logps/rejected": -346.0, | |
| "logps/rejected_bottom_tokens": -14.0625, | |
| "logps/rejected_top_tokens": -0.00084686279296875, | |
| "loss": 0.6832, | |
| "rewards/accuracies": 0.5600000023841858, | |
| "rewards/chosen": -0.0106201171875, | |
| "rewards/margins": 0.0159912109375, | |
| "rewards/rejected": -0.026611328125, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.19627085377821393, | |
| "grad_norm": 2.2125416576513732, | |
| "learning_rate": 4.860112597371772e-07, | |
| "logits/chosen": -1.125, | |
| "logits/rejected": -1.171875, | |
| "logps/chosen": -372.0, | |
| "logps/chosen_bottom_tokens": -14.125, | |
| "logps/chosen_top_tokens": -0.000904083251953125, | |
| "logps/rejected": -328.0, | |
| "logps/rejected_bottom_tokens": -14.0, | |
| "logps/rejected_top_tokens": -0.0009002685546875, | |
| "loss": 0.678, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": -0.017333984375, | |
| "rewards/margins": 0.0361328125, | |
| "rewards/rejected": -0.053466796875, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.19627085377821393, | |
| "eval_logits/chosen": -1.09375, | |
| "eval_logits/rejected": -1.15625, | |
| "eval_logps/chosen": -396.0, | |
| "eval_logps/chosen_bottom_tokens": -14.0625, | |
| "eval_logps/chosen_top_tokens": -0.0008697509765625, | |
| "eval_logps/rejected": -344.0, | |
| "eval_logps/rejected_bottom_tokens": -14.0, | |
| "eval_logps/rejected_top_tokens": -0.0008697509765625, | |
| "eval_loss": 0.6789160370826721, | |
| "eval_rewards/accuracies": 0.5880597233772278, | |
| "eval_rewards/chosen": -0.0274658203125, | |
| "eval_rewards/margins": 0.033203125, | |
| "eval_rewards/rejected": -0.060791015625, | |
| "eval_runtime": 111.5869, | |
| "eval_samples_per_second": 17.923, | |
| "eval_steps_per_second": 0.6, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.21589793915603533, | |
| "grad_norm": 2.438395616681449, | |
| "learning_rate": 4.798048466485017e-07, | |
| "logits/chosen": -1.1015625, | |
| "logits/rejected": -1.109375, | |
| "logps/chosen": -344.0, | |
| "logps/chosen_bottom_tokens": -14.0, | |
| "logps/chosen_top_tokens": -0.000835418701171875, | |
| "logps/rejected": -332.0, | |
| "logps/rejected_bottom_tokens": -14.0, | |
| "logps/rejected_top_tokens": -0.000873565673828125, | |
| "loss": 0.6804, | |
| "rewards/accuracies": 0.5899999737739563, | |
| "rewards/chosen": -0.037109375, | |
| "rewards/margins": 0.02001953125, | |
| "rewards/rejected": -0.05712890625, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.23552502453385674, | |
| "grad_norm": 2.226213549318803, | |
| "learning_rate": 4.725176028314541e-07, | |
| "logits/chosen": -1.109375, | |
| "logits/rejected": -1.1171875, | |
| "logps/chosen": -372.0, | |
| "logps/chosen_bottom_tokens": -14.0, | |
| "logps/chosen_top_tokens": -0.0008544921875, | |
| "logps/rejected": -354.0, | |
| "logps/rejected_bottom_tokens": -14.0, | |
| "logps/rejected_top_tokens": -0.0008544921875, | |
| "loss": 0.6745, | |
| "rewards/accuracies": 0.6399999856948853, | |
| "rewards/chosen": -0.03564453125, | |
| "rewards/margins": 0.0517578125, | |
| "rewards/rejected": -0.08740234375, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.25515210991167814, | |
| "grad_norm": 2.4135162897156706, | |
| "learning_rate": 4.641838020498713e-07, | |
| "logits/chosen": -1.09375, | |
| "logits/rejected": -1.1640625, | |
| "logps/chosen": -408.0, | |
| "logps/chosen_bottom_tokens": -14.1875, | |
| "logps/chosen_top_tokens": -0.000934600830078125, | |
| "logps/rejected": -338.0, | |
| "logps/rejected_bottom_tokens": -14.1875, | |
| "logps/rejected_top_tokens": -0.00092315673828125, | |
| "loss": 0.6674, | |
| "rewards/accuracies": 0.5849999785423279, | |
| "rewards/chosen": -0.0703125, | |
| "rewards/margins": 0.0517578125, | |
| "rewards/rejected": -0.1220703125, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.2747791952894995, | |
| "grad_norm": 2.4502181786024004, | |
| "learning_rate": 4.5484264029156733e-07, | |
| "logits/chosen": -1.1015625, | |
| "logits/rejected": -1.1484375, | |
| "logps/chosen": -386.0, | |
| "logps/chosen_bottom_tokens": -14.125, | |
| "logps/chosen_top_tokens": -0.000812530517578125, | |
| "logps/rejected": -336.0, | |
| "logps/rejected_bottom_tokens": -14.125, | |
| "logps/rejected_top_tokens": -0.00083160400390625, | |
| "loss": 0.6635, | |
| "rewards/accuracies": 0.5900000333786011, | |
| "rewards/chosen": -0.1015625, | |
| "rewards/margins": 0.048828125, | |
| "rewards/rejected": -0.150390625, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.2944062806673209, | |
| "grad_norm": 2.4663119079457614, | |
| "learning_rate": 4.445380514196192e-07, | |
| "logits/chosen": -1.09375, | |
| "logits/rejected": -1.171875, | |
| "logps/chosen": -428.0, | |
| "logps/chosen_bottom_tokens": -14.1875, | |
| "logps/chosen_top_tokens": -0.00087738037109375, | |
| "logps/rejected": -356.0, | |
| "logps/rejected_bottom_tokens": -14.1875, | |
| "logps/rejected_top_tokens": -0.000858306884765625, | |
| "loss": 0.668, | |
| "rewards/accuracies": 0.6350000500679016, | |
| "rewards/chosen": -0.12890625, | |
| "rewards/margins": 0.0673828125, | |
| "rewards/rejected": -0.1962890625, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.3140333660451423, | |
| "grad_norm": 2.455591342132379, | |
| "learning_rate": 4.33318500540218e-07, | |
| "logits/chosen": -1.0859375, | |
| "logits/rejected": -1.1328125, | |
| "logps/chosen": -408.0, | |
| "logps/chosen_bottom_tokens": -14.125, | |
| "logps/chosen_top_tokens": -0.00089263916015625, | |
| "logps/rejected": -368.0, | |
| "logps/rejected_bottom_tokens": -14.0625, | |
| "logps/rejected_top_tokens": -0.000904083251953125, | |
| "loss": 0.6655, | |
| "rewards/accuracies": 0.6050000190734863, | |
| "rewards/chosen": -0.16796875, | |
| "rewards/margins": 0.078125, | |
| "rewards/rejected": -0.24609375, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.3336604514229637, | |
| "grad_norm": 2.436300399124971, | |
| "learning_rate": 4.2123675605892985e-07, | |
| "logits/chosen": -1.078125, | |
| "logits/rejected": -1.1484375, | |
| "logps/chosen": -422.0, | |
| "logps/chosen_bottom_tokens": -14.3125, | |
| "logps/chosen_top_tokens": -0.00101470947265625, | |
| "logps/rejected": -364.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.00098419189453125, | |
| "loss": 0.6585, | |
| "rewards/accuracies": 0.6450000405311584, | |
| "rewards/chosen": -0.1826171875, | |
| "rewards/margins": 0.115234375, | |
| "rewards/rejected": -0.296875, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.35328753680078506, | |
| "grad_norm": 2.5546008416763035, | |
| "learning_rate": 4.0834964149744333e-07, | |
| "logits/chosen": -1.1015625, | |
| "logits/rejected": -1.15625, | |
| "logps/chosen": -416.0, | |
| "logps/chosen_bottom_tokens": -14.25, | |
| "logps/chosen_top_tokens": -0.00093841552734375, | |
| "logps/rejected": -380.0, | |
| "logps/rejected_bottom_tokens": -14.1875, | |
| "logps/rejected_top_tokens": -0.00099945068359375, | |
| "loss": 0.6643, | |
| "rewards/accuracies": 0.6100000143051147, | |
| "rewards/chosen": -0.232421875, | |
| "rewards/margins": 0.06396484375, | |
| "rewards/rejected": -0.296875, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.3729146221786065, | |
| "grad_norm": 2.555290762655567, | |
| "learning_rate": 3.947177682380738e-07, | |
| "logits/chosen": -1.1796875, | |
| "logits/rejected": -1.203125, | |
| "logps/chosen": -378.0, | |
| "logps/chosen_bottom_tokens": -14.25, | |
| "logps/chosen_top_tokens": -0.000789642333984375, | |
| "logps/rejected": -356.0, | |
| "logps/rejected_bottom_tokens": -14.1875, | |
| "logps/rejected_top_tokens": -0.000823974609375, | |
| "loss": 0.6499, | |
| "rewards/accuracies": 0.6450001001358032, | |
| "rewards/chosen": -0.2412109375, | |
| "rewards/margins": 0.11669921875, | |
| "rewards/rejected": -0.357421875, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.39254170755642787, | |
| "grad_norm": 2.7737043586573313, | |
| "learning_rate": 3.804052504529933e-07, | |
| "logits/chosen": -1.1796875, | |
| "logits/rejected": -1.2265625, | |
| "logps/chosen": -392.0, | |
| "logps/chosen_bottom_tokens": -14.375, | |
| "logps/chosen_top_tokens": -0.0008392333984375, | |
| "logps/rejected": -370.0, | |
| "logps/rejected_bottom_tokens": -14.375, | |
| "logps/rejected_top_tokens": -0.00087738037109375, | |
| "loss": 0.645, | |
| "rewards/accuracies": 0.6350000500679016, | |
| "rewards/chosen": -0.271484375, | |
| "rewards/margins": 0.1396484375, | |
| "rewards/rejected": -0.41015625, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.39254170755642787, | |
| "eval_logits/chosen": -1.15625, | |
| "eval_logits/rejected": -1.203125, | |
| "eval_logps/chosen": -422.0, | |
| "eval_logps/chosen_bottom_tokens": -14.375, | |
| "eval_logps/chosen_top_tokens": -0.000911712646484375, | |
| "eval_logps/rejected": -380.0, | |
| "eval_logps/rejected_bottom_tokens": -14.3125, | |
| "eval_logps/rejected_top_tokens": -0.000919342041015625, | |
| "eval_loss": 0.6488671898841858, | |
| "eval_rewards/accuracies": 0.6447761058807373, | |
| "eval_rewards/chosen": -0.287109375, | |
| "eval_rewards/margins": 0.13671875, | |
| "eval_rewards/rejected": -0.423828125, | |
| "eval_runtime": 111.5112, | |
| "eval_samples_per_second": 17.935, | |
| "eval_steps_per_second": 0.601, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.41216879293424924, | |
| "grad_norm": 2.8286672144445277, | |
| "learning_rate": 3.654794035589483e-07, | |
| "logits/chosen": -1.1328125, | |
| "logits/rejected": -1.1640625, | |
| "logps/chosen": -362.0, | |
| "logps/chosen_bottom_tokens": -14.25, | |
| "logps/chosen_top_tokens": -0.00091552734375, | |
| "logps/rejected": -344.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.000926971435546875, | |
| "loss": 0.6512, | |
| "rewards/accuracies": 0.6149999499320984, | |
| "rewards/chosen": -0.298828125, | |
| "rewards/margins": 0.12060546875, | |
| "rewards/rejected": -0.419921875, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.43179587831207067, | |
| "grad_norm": 2.98579141751378, | |
| "learning_rate": 3.5001042761570826e-07, | |
| "logits/chosen": -1.171875, | |
| "logits/rejected": -1.2109375, | |
| "logps/chosen": -414.0, | |
| "logps/chosen_bottom_tokens": -14.5, | |
| "logps/chosen_top_tokens": -0.000762939453125, | |
| "logps/rejected": -398.0, | |
| "logps/rejected_bottom_tokens": -14.375, | |
| "logps/rejected_top_tokens": -0.000743865966796875, | |
| "loss": 0.6507, | |
| "rewards/accuracies": 0.5800000429153442, | |
| "rewards/chosen": -0.333984375, | |
| "rewards/margins": 0.11279296875, | |
| "rewards/rejected": -0.447265625, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.45142296368989204, | |
| "grad_norm": 2.849801650804548, | |
| "learning_rate": 3.34071077157304e-07, | |
| "logits/chosen": -1.171875, | |
| "logits/rejected": -1.2265625, | |
| "logps/chosen": -388.0, | |
| "logps/chosen_bottom_tokens": -14.375, | |
| "logps/chosen_top_tokens": -0.00075531005859375, | |
| "logps/rejected": -354.0, | |
| "logps/rejected_bottom_tokens": -14.3125, | |
| "logps/rejected_top_tokens": -0.000827789306640625, | |
| "loss": 0.6464, | |
| "rewards/accuracies": 0.6299999952316284, | |
| "rewards/chosen": -0.353515625, | |
| "rewards/margins": 0.1337890625, | |
| "rewards/rejected": -0.48828125, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.47105004906771347, | |
| "grad_norm": 3.020709895469043, | |
| "learning_rate": 3.1773631900892204e-07, | |
| "logits/chosen": -1.1484375, | |
| "logits/rejected": -1.1875, | |
| "logps/chosen": -416.0, | |
| "logps/chosen_bottom_tokens": -14.375, | |
| "logps/chosen_top_tokens": -0.000759124755859375, | |
| "logps/rejected": -396.0, | |
| "logps/rejected_bottom_tokens": -14.375, | |
| "logps/rejected_top_tokens": -0.0007781982421875, | |
| "loss": 0.6442, | |
| "rewards/accuracies": 0.6200000047683716, | |
| "rewards/chosen": -0.38671875, | |
| "rewards/margins": 0.1337890625, | |
| "rewards/rejected": -0.51953125, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.49067713444553485, | |
| "grad_norm": 2.801068325901482, | |
| "learning_rate": 3.0108297969883103e-07, | |
| "logits/chosen": -1.1640625, | |
| "logits/rejected": -1.1953125, | |
| "logps/chosen": -426.0, | |
| "logps/chosen_bottom_tokens": -14.4375, | |
| "logps/chosen_top_tokens": -0.0008392333984375, | |
| "logps/rejected": -398.0, | |
| "logps/rejected_bottom_tokens": -14.375, | |
| "logps/rejected_top_tokens": -0.00081634521484375, | |
| "loss": 0.6347, | |
| "rewards/accuracies": 0.64000004529953, | |
| "rewards/chosen": -0.39453125, | |
| "rewards/margins": 0.1865234375, | |
| "rewards/rejected": -0.58203125, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5103042198233563, | |
| "grad_norm": 2.8119914001202835, | |
| "learning_rate": 2.8418938412365013e-07, | |
| "logits/chosen": -1.1640625, | |
| "logits/rejected": -1.203125, | |
| "logps/chosen": -396.0, | |
| "logps/chosen_bottom_tokens": -14.3125, | |
| "logps/chosen_top_tokens": -0.000865936279296875, | |
| "logps/rejected": -372.0, | |
| "logps/rejected_bottom_tokens": -14.1875, | |
| "logps/rejected_top_tokens": -0.000885009765625, | |
| "loss": 0.6381, | |
| "rewards/accuracies": 0.6350000500679016, | |
| "rewards/chosen": -0.373046875, | |
| "rewards/margins": 0.171875, | |
| "rewards/rejected": -0.546875, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.5299313052011776, | |
| "grad_norm": 2.914608701481186, | |
| "learning_rate": 2.671349871664101e-07, | |
| "logits/chosen": -1.1640625, | |
| "logits/rejected": -1.171875, | |
| "logps/chosen": -398.0, | |
| "logps/chosen_bottom_tokens": -14.3125, | |
| "logps/chosen_top_tokens": -0.000782012939453125, | |
| "logps/rejected": -386.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.00077056884765625, | |
| "loss": 0.6315, | |
| "rewards/accuracies": 0.64000004529953, | |
| "rewards/chosen": -0.40625, | |
| "rewards/margins": 0.17578125, | |
| "rewards/rejected": -0.58203125, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.549558390578999, | |
| "grad_norm": 2.974677635397429, | |
| "learning_rate": 2.5e-07, | |
| "logits/chosen": -1.171875, | |
| "logits/rejected": -1.1953125, | |
| "logps/chosen": -438.0, | |
| "logps/chosen_bottom_tokens": -14.3125, | |
| "logps/chosen_top_tokens": -0.000911712646484375, | |
| "logps/rejected": -402.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.0009002685546875, | |
| "loss": 0.6384, | |
| "rewards/accuracies": 0.6600000262260437, | |
| "rewards/chosen": -0.42578125, | |
| "rewards/margins": 0.21484375, | |
| "rewards/rejected": -0.640625, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.5691854759568205, | |
| "grad_norm": 3.4767790428686234, | |
| "learning_rate": 2.3286501283358982e-07, | |
| "logits/chosen": -1.1796875, | |
| "logits/rejected": -1.2421875, | |
| "logps/chosen": -412.0, | |
| "logps/chosen_bottom_tokens": -14.3125, | |
| "logps/chosen_top_tokens": -0.0008392333984375, | |
| "logps/rejected": -376.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.000885009765625, | |
| "loss": 0.632, | |
| "rewards/accuracies": 0.5750000476837158, | |
| "rewards/chosen": -0.4609375, | |
| "rewards/margins": 0.16015625, | |
| "rewards/rejected": -0.62109375, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.5888125613346418, | |
| "grad_norm": 3.0983859451271565, | |
| "learning_rate": 2.1581061587634987e-07, | |
| "logits/chosen": -1.203125, | |
| "logits/rejected": -1.2421875, | |
| "logps/chosen": -428.0, | |
| "logps/chosen_bottom_tokens": -14.3125, | |
| "logps/chosen_top_tokens": -0.00075531005859375, | |
| "logps/rejected": -388.0, | |
| "logps/rejected_bottom_tokens": -14.3125, | |
| "logps/rejected_top_tokens": -0.00079345703125, | |
| "loss": 0.6396, | |
| "rewards/accuracies": 0.5999999642372131, | |
| "rewards/chosen": -0.482421875, | |
| "rewards/margins": 0.162109375, | |
| "rewards/rejected": -0.64453125, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5888125613346418, | |
| "eval_logits/chosen": -1.1875, | |
| "eval_logits/rejected": -1.234375, | |
| "eval_logps/chosen": -438.0, | |
| "eval_logps/chosen_bottom_tokens": -14.375, | |
| "eval_logps/chosen_top_tokens": -0.0007476806640625, | |
| "eval_logps/rejected": -406.0, | |
| "eval_logps/rejected_bottom_tokens": -14.3125, | |
| "eval_logps/rejected_top_tokens": -0.000759124755859375, | |
| "eval_loss": 0.6303857564926147, | |
| "eval_rewards/accuracies": 0.6626865863800049, | |
| "eval_rewards/chosen": -0.451171875, | |
| "eval_rewards/margins": 0.2275390625, | |
| "eval_rewards/rejected": -0.6796875, | |
| "eval_runtime": 111.5027, | |
| "eval_samples_per_second": 17.937, | |
| "eval_steps_per_second": 0.601, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6084396467124632, | |
| "grad_norm": 3.1412458629194835, | |
| "learning_rate": 1.9891702030116897e-07, | |
| "logits/chosen": -1.140625, | |
| "logits/rejected": -1.2421875, | |
| "logps/chosen": -446.0, | |
| "logps/chosen_bottom_tokens": -14.375, | |
| "logps/chosen_top_tokens": -0.00074005126953125, | |
| "logps/rejected": -358.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.0007171630859375, | |
| "loss": 0.6234, | |
| "rewards/accuracies": 0.6949999928474426, | |
| "rewards/chosen": -0.408203125, | |
| "rewards/margins": 0.2451171875, | |
| "rewards/rejected": -0.65625, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.6280667320902846, | |
| "grad_norm": 3.1923082526436986, | |
| "learning_rate": 1.8226368099107792e-07, | |
| "logits/chosen": -1.1640625, | |
| "logits/rejected": -1.2109375, | |
| "logps/chosen": -424.0, | |
| "logps/chosen_bottom_tokens": -14.3125, | |
| "logps/chosen_top_tokens": -0.00080108642578125, | |
| "logps/rejected": -364.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.000843048095703125, | |
| "loss": 0.6241, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -0.42578125, | |
| "rewards/margins": 0.2216796875, | |
| "rewards/rejected": -0.6484375, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.647693817468106, | |
| "grad_norm": 3.064211696764281, | |
| "learning_rate": 1.6592892284269594e-07, | |
| "logits/chosen": -1.1796875, | |
| "logits/rejected": -1.2109375, | |
| "logps/chosen": -408.0, | |
| "logps/chosen_bottom_tokens": -14.375, | |
| "logps/chosen_top_tokens": -0.0007171630859375, | |
| "logps/rejected": -386.0, | |
| "logps/rejected_bottom_tokens": -14.3125, | |
| "logps/rejected_top_tokens": -0.000732421875, | |
| "loss": 0.6224, | |
| "rewards/accuracies": 0.6799999475479126, | |
| "rewards/chosen": -0.431640625, | |
| "rewards/margins": 0.259765625, | |
| "rewards/rejected": -0.69140625, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.6673209028459274, | |
| "grad_norm": 3.1791023826814353, | |
| "learning_rate": 1.4998957238429172e-07, | |
| "logits/chosen": -1.21875, | |
| "logits/rejected": -1.2421875, | |
| "logps/chosen": -408.0, | |
| "logps/chosen_bottom_tokens": -14.3125, | |
| "logps/chosen_top_tokens": -0.000804901123046875, | |
| "logps/rejected": -380.0, | |
| "logps/rejected_bottom_tokens": -14.3125, | |
| "logps/rejected_top_tokens": -0.00077056884765625, | |
| "loss": 0.6204, | |
| "rewards/accuracies": 0.6300000548362732, | |
| "rewards/chosen": -0.5078125, | |
| "rewards/margins": 0.244140625, | |
| "rewards/rejected": -0.75390625, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.6869479882237488, | |
| "grad_norm": 3.295570474728778, | |
| "learning_rate": 1.345205964410517e-07, | |
| "logits/chosen": -1.1953125, | |
| "logits/rejected": -1.21875, | |
| "logps/chosen": -392.0, | |
| "logps/chosen_bottom_tokens": -14.375, | |
| "logps/chosen_top_tokens": -0.000858306884765625, | |
| "logps/rejected": -372.0, | |
| "logps/rejected_bottom_tokens": -14.3125, | |
| "logps/rejected_top_tokens": -0.00089263916015625, | |
| "loss": 0.627, | |
| "rewards/accuracies": 0.5850000381469727, | |
| "rewards/chosen": -0.49609375, | |
| "rewards/margins": 0.2236328125, | |
| "rewards/rejected": -0.71875, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7065750736015701, | |
| "grad_norm": 3.5211819482445184, | |
| "learning_rate": 1.1959474954700665e-07, | |
| "logits/chosen": -1.1796875, | |
| "logits/rejected": -1.21875, | |
| "logps/chosen": -424.0, | |
| "logps/chosen_bottom_tokens": -14.375, | |
| "logps/chosen_top_tokens": -0.00067138671875, | |
| "logps/rejected": -416.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.000675201416015625, | |
| "loss": 0.613, | |
| "rewards/accuracies": 0.7049999833106995, | |
| "rewards/chosen": -0.44140625, | |
| "rewards/margins": 0.234375, | |
| "rewards/rejected": -0.67578125, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.7262021589793916, | |
| "grad_norm": 3.3333877037469026, | |
| "learning_rate": 1.0528223176192615e-07, | |
| "logits/chosen": -1.1953125, | |
| "logits/rejected": -1.234375, | |
| "logps/chosen": -442.0, | |
| "logps/chosen_bottom_tokens": -14.375, | |
| "logps/chosen_top_tokens": -0.00069427490234375, | |
| "logps/rejected": -398.0, | |
| "logps/rejected_bottom_tokens": -14.375, | |
| "logps/rejected_top_tokens": -0.00067901611328125, | |
| "loss": 0.6218, | |
| "rewards/accuracies": 0.6250000596046448, | |
| "rewards/chosen": -0.5234375, | |
| "rewards/margins": 0.2138671875, | |
| "rewards/rejected": -0.73828125, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.745829244357213, | |
| "grad_norm": 3.3039144354882657, | |
| "learning_rate": 9.16503585025567e-08, | |
| "logits/chosen": -1.1953125, | |
| "logits/rejected": -1.21875, | |
| "logps/chosen": -420.0, | |
| "logps/chosen_bottom_tokens": -14.25, | |
| "logps/chosen_top_tokens": -0.00090789794921875, | |
| "logps/rejected": -412.0, | |
| "logps/rejected_bottom_tokens": -14.1875, | |
| "logps/rejected_top_tokens": -0.00104522705078125, | |
| "loss": 0.6279, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -0.5, | |
| "rewards/margins": 0.2041015625, | |
| "rewards/rejected": -0.703125, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.7654563297350343, | |
| "grad_norm": 3.460907844274303, | |
| "learning_rate": 7.876324394107017e-08, | |
| "logits/chosen": -1.15625, | |
| "logits/rejected": -1.203125, | |
| "logps/chosen": -442.0, | |
| "logps/chosen_bottom_tokens": -14.3125, | |
| "logps/chosen_top_tokens": -0.00067901611328125, | |
| "logps/rejected": -418.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.00069427490234375, | |
| "loss": 0.6289, | |
| "rewards/accuracies": 0.6350000500679016, | |
| "rewards/chosen": -0.50390625, | |
| "rewards/margins": 0.2255859375, | |
| "rewards/rejected": -0.7265625, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.7850834151128557, | |
| "grad_norm": 3.2842912290921897, | |
| "learning_rate": 6.668149945978201e-08, | |
| "logits/chosen": -1.1953125, | |
| "logits/rejected": -1.2265625, | |
| "logps/chosen": -440.0, | |
| "logps/chosen_bottom_tokens": -14.3125, | |
| "logps/chosen_top_tokens": -0.00070953369140625, | |
| "logps/rejected": -420.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.000732421875, | |
| "loss": 0.6102, | |
| "rewards/accuracies": 0.6700000166893005, | |
| "rewards/chosen": -0.482421875, | |
| "rewards/margins": 0.291015625, | |
| "rewards/rejected": -0.7734375, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7850834151128557, | |
| "eval_logits/chosen": -1.1875, | |
| "eval_logits/rejected": -1.234375, | |
| "eval_logps/chosen": -444.0, | |
| "eval_logps/chosen_bottom_tokens": -14.3125, | |
| "eval_logps/chosen_top_tokens": -0.00067138671875, | |
| "eval_logps/rejected": -414.0, | |
| "eval_logps/rejected_bottom_tokens": -14.25, | |
| "eval_logps/rejected_top_tokens": -0.00066375732421875, | |
| "eval_loss": 0.6267920136451721, | |
| "eval_rewards/accuracies": 0.6567164063453674, | |
| "eval_rewards/chosen": -0.50390625, | |
| "eval_rewards/margins": 0.2578125, | |
| "eval_rewards/rejected": -0.76171875, | |
| "eval_runtime": 111.5791, | |
| "eval_samples_per_second": 17.925, | |
| "eval_steps_per_second": 0.6, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8047105004906772, | |
| "grad_norm": 3.3007954730404303, | |
| "learning_rate": 5.546194858038072e-08, | |
| "logits/chosen": -1.171875, | |
| "logits/rejected": -1.21875, | |
| "logps/chosen": -416.0, | |
| "logps/chosen_bottom_tokens": -14.3125, | |
| "logps/chosen_top_tokens": -0.000766754150390625, | |
| "logps/rejected": -374.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.000827789306640625, | |
| "loss": 0.6227, | |
| "rewards/accuracies": 0.6699999570846558, | |
| "rewards/chosen": -0.51171875, | |
| "rewards/margins": 0.28515625, | |
| "rewards/rejected": -0.796875, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.8243375858684985, | |
| "grad_norm": 3.9743687860867185, | |
| "learning_rate": 4.5157359708432626e-08, | |
| "logits/chosen": -1.1953125, | |
| "logits/rejected": -1.2265625, | |
| "logps/chosen": -394.0, | |
| "logps/chosen_bottom_tokens": -14.25, | |
| "logps/chosen_top_tokens": -0.00075531005859375, | |
| "logps/rejected": -412.0, | |
| "logps/rejected_bottom_tokens": -14.1875, | |
| "logps/rejected_top_tokens": -0.0007476806640625, | |
| "loss": 0.6205, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -0.474609375, | |
| "rewards/margins": 0.234375, | |
| "rewards/rejected": -0.70703125, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.8439646712463199, | |
| "grad_norm": 3.1969688623984633, | |
| "learning_rate": 3.581619795012874e-08, | |
| "logits/chosen": -1.1796875, | |
| "logits/rejected": -1.1875, | |
| "logps/chosen": -400.0, | |
| "logps/chosen_bottom_tokens": -14.1875, | |
| "logps/chosen_top_tokens": -0.000782012939453125, | |
| "logps/rejected": -404.0, | |
| "logps/rejected_bottom_tokens": -14.1875, | |
| "logps/rejected_top_tokens": -0.000743865966796875, | |
| "loss": 0.6208, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -0.474609375, | |
| "rewards/margins": 0.267578125, | |
| "rewards/rejected": -0.7421875, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.8635917566241413, | |
| "grad_norm": 3.705663203159775, | |
| "learning_rate": 2.748239716854589e-08, | |
| "logits/chosen": -1.2109375, | |
| "logits/rejected": -1.1953125, | |
| "logps/chosen": -424.0, | |
| "logps/chosen_bottom_tokens": -14.1875, | |
| "logps/chosen_top_tokens": -0.000728607177734375, | |
| "logps/rejected": -420.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.000751495361328125, | |
| "loss": 0.6398, | |
| "rewards/accuracies": 0.5849999785423279, | |
| "rewards/chosen": -0.51171875, | |
| "rewards/margins": 0.1435546875, | |
| "rewards/rejected": -0.65625, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.8832188420019627, | |
| "grad_norm": 3.9792023056235455, | |
| "learning_rate": 2.0195153351498323e-08, | |
| "logits/chosen": -1.1796875, | |
| "logits/rejected": -1.2109375, | |
| "logps/chosen": -432.0, | |
| "logps/chosen_bottom_tokens": -14.375, | |
| "logps/chosen_top_tokens": -0.000732421875, | |
| "logps/rejected": -420.0, | |
| "logps/rejected_bottom_tokens": -14.3125, | |
| "logps/rejected_top_tokens": -0.000705718994140625, | |
| "loss": 0.611, | |
| "rewards/accuracies": 0.6149999499320984, | |
| "rewards/chosen": -0.53125, | |
| "rewards/margins": 0.2421875, | |
| "rewards/rejected": -0.7734375, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9028459273797841, | |
| "grad_norm": 3.598443005581659, | |
| "learning_rate": 1.3988740262822846e-08, | |
| "logits/chosen": -1.1953125, | |
| "logits/rejected": -1.203125, | |
| "logps/chosen": -428.0, | |
| "logps/chosen_bottom_tokens": -14.375, | |
| "logps/chosen_top_tokens": -0.00067138671875, | |
| "logps/rejected": -410.0, | |
| "logps/rejected_bottom_tokens": -14.3125, | |
| "logps/rejected_top_tokens": -0.000667572021484375, | |
| "loss": 0.6138, | |
| "rewards/accuracies": 0.6349999904632568, | |
| "rewards/chosen": -0.490234375, | |
| "rewards/margins": 0.216796875, | |
| "rewards/rejected": -0.70703125, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.9224730127576055, | |
| "grad_norm": 3.423571391469107, | |
| "learning_rate": 8.892348244137788e-09, | |
| "logits/chosen": -1.1875, | |
| "logits/rejected": -1.2421875, | |
| "logps/chosen": -474.0, | |
| "logps/chosen_bottom_tokens": -14.5625, | |
| "logps/chosen_top_tokens": -0.000675201416015625, | |
| "logps/rejected": -444.0, | |
| "logps/rejected_bottom_tokens": -14.375, | |
| "logps/rejected_top_tokens": -0.000652313232421875, | |
| "loss": 0.6106, | |
| "rewards/accuracies": 0.6299999952316284, | |
| "rewards/chosen": -0.494140625, | |
| "rewards/margins": 0.271484375, | |
| "rewards/rejected": -0.765625, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.9421000981354269, | |
| "grad_norm": 3.1667123948106584, | |
| "learning_rate": 4.929946925231076e-09, | |
| "logits/chosen": -1.1328125, | |
| "logits/rejected": -1.171875, | |
| "logps/chosen": -410.0, | |
| "logps/chosen_bottom_tokens": -14.375, | |
| "logps/chosen_top_tokens": -0.0007476806640625, | |
| "logps/rejected": -412.0, | |
| "logps/rejected_bottom_tokens": -14.25, | |
| "logps/rejected_top_tokens": -0.00095367431640625, | |
| "loss": 0.6203, | |
| "rewards/accuracies": 0.6049999594688416, | |
| "rewards/chosen": -0.48046875, | |
| "rewards/margins": 0.1953125, | |
| "rewards/rejected": -0.67578125, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.9617271835132483, | |
| "grad_norm": 3.5902417143779024, | |
| "learning_rate": 2.1201724887858484e-09, | |
| "logits/chosen": -1.1640625, | |
| "logits/rejected": -1.171875, | |
| "logps/chosen": -422.0, | |
| "logps/chosen_bottom_tokens": -14.25, | |
| "logps/chosen_top_tokens": -0.00072479248046875, | |
| "logps/rejected": -412.0, | |
| "logps/rejected_bottom_tokens": -14.1875, | |
| "logps/rejected_top_tokens": -0.000743865966796875, | |
| "loss": 0.6235, | |
| "rewards/accuracies": 0.5949999690055847, | |
| "rewards/chosen": -0.5390625, | |
| "rewards/margins": 0.2265625, | |
| "rewards/rejected": -0.765625, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.9813542688910697, | |
| "grad_norm": 3.3154898943344704, | |
| "learning_rate": 4.762400196664518e-10, | |
| "logits/chosen": -1.1484375, | |
| "logits/rejected": -1.1953125, | |
| "logps/chosen": -428.0, | |
| "logps/chosen_bottom_tokens": -14.375, | |
| "logps/chosen_top_tokens": -0.000621795654296875, | |
| "logps/rejected": -388.0, | |
| "logps/rejected_bottom_tokens": -14.3125, | |
| "logps/rejected_top_tokens": -0.0006256103515625, | |
| "loss": 0.6084, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -0.578125, | |
| "rewards/margins": 0.1787109375, | |
| "rewards/rejected": -0.75390625, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9813542688910697, | |
| "eval_logits/chosen": -1.1953125, | |
| "eval_logits/rejected": -1.2421875, | |
| "eval_logps/chosen": -446.0, | |
| "eval_logps/chosen_bottom_tokens": -14.375, | |
| "eval_logps/chosen_top_tokens": -0.000743865966796875, | |
| "eval_logps/rejected": -416.0, | |
| "eval_logps/rejected_bottom_tokens": -14.3125, | |
| "eval_logps/rejected_top_tokens": -0.0007476806640625, | |
| "eval_loss": 0.6259472370147705, | |
| "eval_rewards/accuracies": 0.6567164659500122, | |
| "eval_rewards/chosen": -0.5234375, | |
| "eval_rewards/margins": 0.26171875, | |
| "eval_rewards/rejected": -0.78515625, | |
| "eval_runtime": 111.4505, | |
| "eval_samples_per_second": 17.945, | |
| "eval_steps_per_second": 0.601, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9990186457311089, | |
| "step": 509, | |
| "total_flos": 0.0, | |
| "train_loss": 0.6464882252961105, | |
| "train_runtime": 8284.9703, | |
| "train_samples_per_second": 7.379, | |
| "train_steps_per_second": 0.061 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 509, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 0.0, | |
| "train_batch_size": 5, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |