diff --git a/evaluation_results/2024-12-21_13-57-31/results.md b/evaluation_results/2024-12-21_13-57-31/results.md
index 3be0a9c..1923c0d 100644
--- a/evaluation_results/2024-12-21_13-57-31/results.md
+++ b/evaluation_results/2024-12-21_13-57-31/results.md
@@ -4,9 +4,7 @@ There are 4 scenarios and 4 test cases with 3 attempts (48 total tests).
 ## Test: blank_math
 
 ### claude_sonnet_latest_with_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/claude_sonnet_latest_with_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluations/blank_math/input.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/claude_sonnet_latest_with_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluations/blank_math/input.png' border=1 width=200 />
 
 ```
 10
@@ -15,36 +13,22 @@ There are 4 scenarios and 4 test cases with 3 attempts (48 total tests).
 <img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/claude_sonnet_latest_with_seg/3/merged-output.png' border=1 width=200 />
 
 ### gpt-4o-mini_no_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/gpt-4o-mini_no_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/gpt-4o-mini_no_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/gpt-4o-mini_no_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/gpt-4o-mini_no_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/gpt-4o-mini_no_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/gpt-4o-mini_no_seg/3/merged-output.png' border=1 width=200 />
 
 ### gpt-4o_with_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/gpt-4o_with_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/gpt-4o_with_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluations/blank_math/input.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/gpt-4o_with_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/gpt-4o_with_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluations/blank_math/input.png' border=1 width=200 />
 
 ```
 10
 ```
 
 ### claude_sonnet_latest_no_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/claude_sonnet_latest_no_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/claude_sonnet_latest_no_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/claude_sonnet_latest_no_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/claude_sonnet_latest_no_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/claude_sonnet_latest_no_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/blank_math/claude_sonnet_latest_no_seg/3/merged-output.png' border=1 width=200 />
 
 ## Test: tic_tac_toe_1
 
 ### claude_sonnet_latest_with_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/claude_sonnet_latest_with_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluations/tic_tac_toe_1/input.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/claude_sonnet_latest_with_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluations/tic_tac_toe_1/input.png' border=1 width=200 />
 
 ```
 Your turn! Place an O anywhere you'd like.
@@ -53,83 +37,39 @@ Your turn! Place an O anywhere you'd like.
 <img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/claude_sonnet_latest_with_seg/3/merged-output.png' border=1 width=200 />
 
 ### gpt-4o-mini_no_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/gpt-4o-mini_no_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/gpt-4o-mini_no_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/gpt-4o-mini_no_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/gpt-4o-mini_no_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/gpt-4o-mini_no_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/gpt-4o-mini_no_seg/3/merged-output.png' border=1 width=200 />
 
 ### gpt-4o_with_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/gpt-4o_with_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/gpt-4o_with_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/gpt-4o_with_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/gpt-4o_with_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/gpt-4o_with_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/gpt-4o_with_seg/3/merged-output.png' border=1 width=200 />
 
 ### claude_sonnet_latest_no_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/claude_sonnet_latest_no_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/claude_sonnet_latest_no_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/claude_sonnet_latest_no_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/claude_sonnet_latest_no_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/claude_sonnet_latest_no_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/tic_tac_toe_1/claude_sonnet_latest_no_seg/3/merged-output.png' border=1 width=200 />
 
 ## Test: x_in_box
 
 ### claude_sonnet_latest_with_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/claude_sonnet_latest_with_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/claude_sonnet_latest_with_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/claude_sonnet_latest_with_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/claude_sonnet_latest_with_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/claude_sonnet_latest_with_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/claude_sonnet_latest_with_seg/3/merged-output.png' border=1 width=200 />
 
 ### gpt-4o-mini_no_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/gpt-4o-mini_no_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/gpt-4o-mini_no_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/gpt-4o-mini_no_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/gpt-4o-mini_no_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/gpt-4o-mini_no_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/gpt-4o-mini_no_seg/3/merged-output.png' border=1 width=200 />
 
 ### gpt-4o_with_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/gpt-4o_with_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/gpt-4o_with_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/gpt-4o_with_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/gpt-4o_with_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/gpt-4o_with_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/gpt-4o_with_seg/3/merged-output.png' border=1 width=200 />
 
 ### claude_sonnet_latest_no_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/claude_sonnet_latest_no_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/claude_sonnet_latest_no_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/claude_sonnet_latest_no_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/claude_sonnet_latest_no_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/claude_sonnet_latest_no_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_box/claude_sonnet_latest_no_seg/3/merged-output.png' border=1 width=200 />
 
 ## Test: x_in_boxes
 
 ### claude_sonnet_latest_with_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/claude_sonnet_latest_with_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/claude_sonnet_latest_with_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/claude_sonnet_latest_with_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/claude_sonnet_latest_with_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/claude_sonnet_latest_with_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/claude_sonnet_latest_with_seg/3/merged-output.png' border=1 width=200 />
 
 ### gpt-4o-mini_no_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/gpt-4o-mini_no_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/gpt-4o-mini_no_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/gpt-4o-mini_no_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/gpt-4o-mini_no_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/gpt-4o-mini_no_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/gpt-4o-mini_no_seg/3/merged-output.png' border=1 width=200 />
 
 ### gpt-4o_with_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/gpt-4o_with_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/gpt-4o_with_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/gpt-4o_with_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/gpt-4o_with_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/gpt-4o_with_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/gpt-4o_with_seg/3/merged-output.png' border=1 width=200 />
 
 ### claude_sonnet_latest_no_seg
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/claude_sonnet_latest_no_seg/1/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/claude_sonnet_latest_no_seg/2/merged-output.png' border=1 width=200 />
-
-<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/claude_sonnet_latest_no_seg/3/merged-output.png' border=1 width=200 />
+<img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/claude_sonnet_latest_no_seg/1/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/claude_sonnet_latest_no_seg/2/merged-output.png' border=1 width=200 /> <img src='../../evaluation_results/2024-12-21_13-57-31/x_in_boxes/claude_sonnet_latest_no_seg/3/merged-output.png' border=1 width=200 />