changed structure

eval/yaml/r1/algebra.yaml

@@ -0,0 +1,13 @@
+model: deepseek/deepseek-r1
+category: algebra
+datasets:
+  - intermediate_integration
+  - polynomial_equations
+  - polynomial_multiplication
+  - simple_equations
+  - simple_integration
+  - complex_arithmetic
+eval_dir: eval/r1
+dataset_size: 50
+dataset_seed: 42
+developer_role: system

eval/yaml/r1/algorithmic.yaml

@@ -0,0 +1,34 @@
+model: deepseek/deepseek-r1
+category: algorithmic
+datasets:
+  - ab
+  -  base_conversion
+  -  binary_matrix
+  -  caesar_cipher
+  -  count_primes
+  -  game_of_life
+  -  graph_color
+  -  group_anagrams
+  -  isomorphic_strings
+  -  letter_counting
+  -  letter_jumble
+  -  manipulate_matrix
+  -  number_filtering
+  -  number_sorting
+  -  palindrome
+  -  pool_matrix
+  -  ransom_note
+  -  rotate_matrix
+  -  sentence_reordering
+  -  spell_backward
+  -  spiral_matrix
+  -  string_insertion
+  -  string_manipulation
+  -  string_synthesis
+  -  word_ladder
+  -  word_sequence_reversal
+  -  word_sorting
+eval_dir: eval/r1
+dataset_size: 50
+dataset_seed: 45
+developer_role: system

eval/yaml/r1/cognition.yaml

@@ -0,0 +1,11 @@
+model: deepseek/deepseek-r1
+category: cognition
+datasets:
+  -  color_cube_rotation
+  -  figlet_font
+  -  number_sequence
+  -  rubiks_cube
+eval_dir: eval/r1
+dataset_size: 50
+dataset_seed: 42
+developer_role: system

eval/yaml/r1/logic.yaml

@@ -0,0 +1,11 @@
+model: deepseek/deepseek-r1
+category: logic
+datasets:
+  - propositional_logic
+  - self_reference
+  - syllogism
+  - zebra_puzzles
+eval_dir: eval/r1
+dataset_size: 50
+dataset_seed: 42
+developer_role: system