MathCAMPS Logo

CodeLlama 34B

Performance on individual Common Core standards, grouped by grade level.
IFUP Acc. = Incremental Followup Accuracy, CFUP Acc. = Counterfactual Followup Accuracy, Total FUPs Seen = Number of followup questions a model sees, since a model only sees followups if it answers the main question correctly.

Grade K

Standard Overall Acc. IFUP Acc. CFUP Acc. Total FUPs
K.OA.A.4 0.94 - - -
K.NBT.A.1 0.92 - - -
K.OA.A.5 0.86 0.83 0.96 41
K.CC.C.7 0.88 - 0.93 84

Grade 1

Standard Overall Acc. IFUP Acc. CFUP Acc. Total FUPs
1.OA.A.2 0.96 1.00 1.00 26
1.OA.A.1 0.96 0.89 1.00 20
1.OA.D.8 0.90 - - -

Grade 2

Standard Overall Acc. IFUP Acc. CFUP Acc. Total FUPs
2.NBT.B.7 0.88 0.94 0.95 38
2.MD.B.5 0.93 0.89 0.91 20
2.NBT.B.6 0.88 1.00 0.87 38
2.OA.A.1 0.91 - - -
2.NBT.B.5 0.93 0.73 1.00 39
2.MD.C.8 0.96 0.90 0.94 53

Grade 3

Standard Overall Acc. IFUP Acc. CFUP Acc. Total FUPs
3.OA.D.8 0.94 1.00 1.00 30
3.MD.D.8-quadrilateral 0.76 - - -
3.OA.A.4 0.90 - - -
3.MD.D.8-polygon 0.49 - - -
3.MD.D.8-triangle 0.90 - - -
3.NBT.A.2 0.93 0.83 1.00 39
3.OA.A.3 0.94 1.00 0.81 37
3.OA.C.7 0.93 0.87 0.97 44

Grade 4

Standard Overall Acc. IFUP Acc. CFUP Acc. Total FUPs
4.NBT.B.4 0.90 1.00 1.00 29
4.OA.B.4 0.23 - - -
4.MD.A.2-decimal 0.68 0.85 0.88 29
4.MD.A.3 0.61 - 0.72 25
4.MD.A.2-fraction 0.27 0.56 0.80 19
4.NBT.B.6 0.33 - 0.67 6
4.NF.A.2 0.59 - 0.86 14
4.NBT.B.5 0.50 - 0.62 24
4.OA.A.3 0.51 0.75 1.00 11

Grade 5

Standard Overall Acc. IFUP Acc. CFUP Acc. Total FUPs
5.OA.A.1 0.59 0.50 0.77 17
5.NBT.B.6 0.68 - - 1
5.NF.A.1 0.00 - - -
5.NBT.B.5 0.25 - 0.75 4
5.NF.B.4 0.48 0.27 0.77 28
5.NBT.B.7 0.59 0.75 1.00 9
5.NF.A.2 0.08 - - 5

Grade 6

Standard Overall Acc. IFUP Acc. CFUP Acc. Total FUPs
6.EE.B.7 0.90 - - -
6.NS.B.2 0.46 - - -
6.EE.A.1 1.00 - 1.00 22
6.NS.B.3 0.43 - 0.50 7

Grade 7

Standard Overall Acc. IFUP Acc. CFUP Acc. Total FUPs
7.NS.A.1-fraction 0.02 1.00 - 2
7.NS.A.2 0.23 0.33 0.33 12
7.NS.A.1-decimal 0.88 0.84 0.82 59
7.NS.A.3-fraction 0.04 - - -
7.NS.A.3-decimal 0.67 1.00 0.83 12

Grade 8

Standard Overall Acc. IFUP Acc. CFUP Acc. Total FUPs
8.EE.A.2 0.64 - - -
8.EE.C.8 0.00 - - -
8.EE.C.7 0.26 - - -