Tool A/B Spot Check - Base vs Tuned (2026-05-30)

Generated: 2026-06-01 01:28:55Z

Scope

• Manual spot-check battery: 5 prompts per tool across 12 Athena tools (60 total A/B pairs)
• Base model: Qwen/Qwen3.5-4B
• Tuned adapter: D:\temp\memorysmith-training\runs\distilled-all-cat-20260531-v7-clean-optimized\adapter
• Raw results JSON: Data/Pages/research/training/tool-ab-spotcheck-20260531-v7-clean-optimized.data.json

Headline Metrics

Per-Tool Results

Notable Improvements

• memorysmith_unified_search-2 (memorysmith_unified_search): base(match=False, env=False, pred=None) -> tuned(match=False, env=True, pred=memorysmith_documentation_search); prompt="find docs for training harness"
• memorysmith_unified_search-3 (memorysmith_unified_search): base(match=False, env=False, pred=None) -> tuned(match=False, env=True, pred=memorysmith_search); prompt="lookup wiki notes about chat template"
• memorysmith_unified_search-4 (memorysmith_unified_search): base(match=False, env=False, pred=None) -> tuned(match=False, env=True, pred=memorysmith_search); prompt="search for model profile defaults"
• memorysmith_hybrid_search-2 (memorysmith_hybrid_search): base(match=False, env=False, pred=None) -> tuned(match=True, env=True, pred=memorysmith_hybrid_search); prompt="hybrid search for source bundle auth behavior"
• memorysmith_semantic_search-2 (memorysmith_semantic_search): base(match=False, env=False, pred=None) -> tuned(match=False, env=True, pred=memorysmith_search); prompt="semantic search for pages lock navigation regression"
• memorysmith_semantic_search-3 (memorysmith_semantic_search): base(match=False, env=False, pred=None) -> tuned(match=True, env=True, pred=memorysmith_semantic_search); prompt="semantic search for sidebar collapse standard"
• memorysmith_search-3 (memorysmith_search): base(match=False, env=False, pred=None) -> tuned(match=True, env=True, pred=memorysmith_search); prompt="find literal text RequestGuard"
• memorysmith_get-1 (memorysmith_get): base(match=False, env=False, pred=None) -> tuned(match=False, env=True, pred=memory_get); prompt="show memory mem_project_001"
• memorysmith_page_search-2 (memorysmith_page_search): base(match=False, env=False, pred=None) -> tuned(match=False, env=True, pred=memorysmith_search); prompt="search wiki pages for training harness"
• memorysmith_page_search-3 (memorysmith_page_search): base(match=False, env=False, pred=None) -> tuned(match=False, env=True, pred=memorysmith_page_get); prompt="find page notes about vector search"
• memorysmith_page_search-4 (memorysmith_page_search): base(match=False, env=False, pred=None) -> tuned(match=False, env=True, pred=memorysmith_search); prompt="search pages for request guard middleware"
• memorysmith_page_search-5 (memorysmith_page_search): base(match=False, env=False, pred=None) -> tuned(match=False, env=True, pred=memorysmith_search); prompt="find pages about maintenance revision cycle"

Notable Regressions

• None

Persistent Failures (Both Models)

• memorysmith_unified_search-1 (memorysmith_unified_search): base pred=None, tuned pred=None, baseErr=No JSON object found, tunedErr=No JSON object found
• memorysmith_unified_search-2 (memorysmith_unified_search): base pred=None, tuned pred=memorysmith_documentation_search, baseErr=No JSON object found, tunedErr=None
• memorysmith_unified_search-3 (memorysmith_unified_search): base pred=None, tuned pred=memorysmith_search, baseErr=No JSON object found, tunedErr=None
• memorysmith_unified_search-4 (memorysmith_unified_search): base pred=None, tuned pred=memorysmith_search, baseErr=No JSON object found, tunedErr=None
• memorysmith_unified_search-5 (memorysmith_unified_search): base pred=None, tuned pred=None, baseErr=No JSON object found, tunedErr=No JSON object found
• memorysmith_hybrid_search-1 (memorysmith_hybrid_search): base pred=None, tuned pred=None, baseErr=No JSON object found, tunedErr=No JSON object found
• memorysmith_hybrid_search-3 (memorysmith_hybrid_search): base pred=None, tuned pred=None, baseErr=No JSON object found, tunedErr=No JSON object found
• memorysmith_hybrid_search-4 (memorysmith_hybrid_search): base pred=None, tuned pred=None, baseErr=No JSON object found, tunedErr=No JSON object found
• memorysmith_hybrid_search-5 (memorysmith_hybrid_search): base pred=None, tuned pred=None, baseErr=No JSON object found, tunedErr=No JSON object found
• memorysmith_semantic_search-1 (memorysmith_semantic_search): base pred=None, tuned pred=None, baseErr=No JSON object found, tunedErr=No JSON object found
• memorysmith_semantic_search-2 (memorysmith_semantic_search): base pred=None, tuned pred=memorysmith_search, baseErr=No JSON object found, tunedErr=None
• memorysmith_semantic_search-4 (memorysmith_semantic_search): base pred=None, tuned pred=None, baseErr=No JSON object found, tunedErr=No JSON object found
• memorysmith_semantic_search-5 (memorysmith_semantic_search): base pred=None, tuned pred=None, baseErr=No JSON object found, tunedErr=No JSON object found
• memorysmith_search-1 (memorysmith_search): base pred=None, tuned pred=None, baseErr=No JSON object found, tunedErr=No JSON object found
• memorysmith_search-2 (memorysmith_search): base pred=None, tuned pred=None, baseErr=No JSON object found, tunedErr=No JSON object found

Representative Output Snippets

memorysmith_unified_search-1

• Prompt: search the wiki for kv cache options
• Base: [think] Okay, the user is asking me to search the wiki for information about kv cache options. I need to remember that I'm Athena, MemorySmith's local wiki assistant. When a search or retrieval action is requested, I should respond with ...
• Tuned: The user is asking me to search the wiki for information about KV cache options. This is a search request, so I should use the memorysmith_search tool. I'll include the query as the search term. [/think] {"toolCalls":[{"name":"memorysmit...

memorysmith_unified_search-2

• Prompt: find docs for training harness
• Base: assistant [think] Okay, the user is asking for documentation on the training harness. I need to figure out how to respond as Athena, the local wiki assistant. First, I should check if there's a specific tool I need to use for retrieving ...
• Tuned: The user is asking for documentation about the training harness. This is a straightforward documentation lookup request. I should use the memorysmith_documentation_search tool to find relevant documentation pages. [/think] {"toolCalls":[...

memorysmith_unified_search-3

• Prompt: lookup wiki notes about chat template
• Base: assistant [think] Okay, the user is asking me to look up wiki notes about chat templates. I need to check if there's any information in the MemorySmith's wiki about chat templates. Since I'm Athena, the local wiki assistant, I should use...
• Tuned: The user is asking for a search for the chat template. I should use the memorysmith_search tool to look for this. [/think] {"toolCalls":[{"name":"memorysmith_search","arguments":{"query":"chat template","maxResults":5}}]} assistant [thin...

memorysmith_unified_search-4

• Prompt: search for model profile defaults
• Base: assistant [think] Okay, the user is asking me to search for model profile defaults. I need to figure out what they mean by that. Model profiles could refer to different things depending on the context. Maybe they're talking about machine...
• Tuned: The user is asking for information about model profile defaults. This is a search request, so I should use the memorysmith_search tool to find relevant memories. I'll include the query as the search term. [/think] {"toolCalls":[{"name":"...

memorysmith_unified_search-5

• Prompt: find references about code search eta
• Base: assistant [think] Okay, the user is asking for references about code search ETA. First, I need to figure out what they mean by "code search eta". ETA usually stands for Estimated Time of Arrival, but in the context of code search, that d...
• Tuned: The user is asking for information about code search ETA. This is a request for information from the knowledge base, so I should use the memorysmith_search tool to look for relevant memories. I'll include the query as the search term. [/...

memorysmith_hybrid_search-1

• Prompt: run a hybrid search for request guard middleware
• Base: [think] Okay, the user is asking me to run a hybrid search for "request guard middleware". I need to figure out how to handle this. First, I remember that as Athena, I'm supposed to respond with a JSON object when a search/retrieval acti...
• Tuned: The user is asking for a hybrid search for "request guard middleware". This is a search request, so I should use the memorysmith_search tool with the hybrid mode. I'll include the query as the search term. [/think] {"toolCalls":[{"name":...

memorysmith_hybrid_search-2

• Prompt: hybrid search for source bundle auth behavior
• Base: [think] Okay, the user is asking about hybrid search for source bundle auth behavior. Let me start by recalling what I know about hybrid search. Hybrid search usually combines different search methods, like keyword and vector search, to ...
• Tuned: The user is asking for a hybrid search query about source bundle authentication behavior. I should respond with a single JSON object containing a tool call for the hybrid search tool. [/think] {"toolCalls":[{"name":"memorysmith_hybrid_se...

memorysmith_hybrid_search-3

• Prompt: use hybrid search to find onnx semantic search notes
• Base: Okay, the user is asking me to use hybrid search to find ONNX semantic search notes. First, I need to understand what they're looking for. Hybrid search typically combines different search methods, like keyword and semantic search, to re...
• Tuned: The user is asking for a hybrid search to find notes about ONNX semantic search. This is a search request, so I should use the memorysmith_search tool with the hybrid mode. I'll include the query about ONNX semantic search in the search ...