Upload configuration file

config/difficulty_sorting_high_seed_math.yaml

@@ -0,0 +1,70 @@
+operators: 
+  - id: load_preexisting
+    config:
+      type: load_preexisting
+      framework_name: base_get_difficulty_seed_math
+  - id: select_difficulty_9
+    config:
+      type: function
+      function: data_strategies.commons.select_rows
+      function_config:
+        conditions:
+          difficulty: 9
+    input_ids:
+      - load_preexisting
+  - id: select_difficulty_10
+    config:
+      type: function
+      function: data_strategies.commons.select_rows
+      function_config:
+        conditions:
+          difficulty: 10
+    input_ids:
+      - load_preexisting
+  - id: select_difficulty_8
+    config:
+      type: function
+      function: data_strategies.commons.select_rows
+      function_config:
+        conditions:
+          difficulty: 8
+    input_ids:
+      - load_preexisting
+  - id: sample_dataset
+    config:
+      type: function
+      function: data_strategies.commons.uniform_sample_limited
+      function_config:
+        num_samples: 3 # 1.25% of 10000 target from stage 1
+    input_ids:
+      - select_difficulty_10
+      - select_difficulty_9
+      - select_difficulty_8
+  # - id: annotate
+  #   config:
+  #     type: completions
+  #     map: chat
+  #     map_config:
+  #       user_message_column: instruction_seed
+  #       output_column: r1_distill_70b_response
+  #     model: "together_ai/deepseek-ai/DeepSeek-R1-Distill-Llama-70B"
+  #     batch: False
+  #     temperature: 1.0
+  #     backend_params:
+  #       max_requests_per_minute: 500
+  #       max_tokens_per_minute: 1_000_000
+  #       invalid_finish_reasons: ['content_filter']
+  #     generation_params:
+  #       max_tokens: 32_000
+  #   input_ids:
+  #     - sample_dataset
+  # - id: convert_to_sharegpt
+  #   config:
+  #     type: function
+  #     function: data_strategies.commons.convert_instruction_response_to_sharegpt
+  #     function_config:
+  #       input_instruction_column: instruction_seed
+  #       input_response_column: r1_distill_70b_response
+  #       output_sharegpt_column: conversations
+  #   input_ids:
+  #     - annotate