feat: native-bakeoff scaffold — Ollama JSON vs native-token tool-calling
Three-arm harness under scripts/native-bakeoff/: - arm A: /api/chat with JSON tools (current default) - arm B: /api/generate raw:true with canonical HF jinja template rendered directly - arm C: google-deepmind/gemma JAX ToolSampler (env-gated, JAX required) Interim finding from A+B sweep on matt-strix gemma4:26b Q4: Ollama's bidirectional JSON↔native tool-call translator is faithful. The "long" multi-tool task produces identical behavior (7 steps / 6 tools) on both arms. Earlier arm-B parser bug that looked like a divergence was a harness issue: preserving the model's <|channel>thought\n<channel|> prefix as assistant content tripped the jinja template's tool_response-following conditional, appending a spurious <turn|>\n that corrupted the next step's prompt. Fixed by dropping the channel prefix on the assistant message. Arm C left as scaffolded-but-not-run — the JAX/bf16 reference path would answer "does the GGUF runtime diverge from DeepMind's implementation" but requires a separate env with the `gemma` PyPI package. Parked pending SDXL eviction or vast-h100 session. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
@@ -0,0 +1,55 @@
|
||||
{
|
||||
"arm": "ollama-json",
|
||||
"model": "gemma4:26b",
|
||||
"num_ctx": 8192,
|
||||
"num_predict": 2048,
|
||||
"started_at": 1776600280.7781782,
|
||||
"turns": [
|
||||
{
|
||||
"step": 1,
|
||||
"elapsed_s": 0.87,
|
||||
"prompt_eval_count": 1322,
|
||||
"eval_count": 30,
|
||||
"content_len": 0,
|
||||
"tool_call_count": 1,
|
||||
"history_chars_before_append": 2697
|
||||
},
|
||||
{
|
||||
"step": 2,
|
||||
"elapsed_s": 0.92,
|
||||
"prompt_eval_count": 1501,
|
||||
"eval_count": 27,
|
||||
"content_len": 0,
|
||||
"tool_call_count": 1,
|
||||
"history_chars_before_append": 3306
|
||||
},
|
||||
{
|
||||
"step": 3,
|
||||
"elapsed_s": 1.41,
|
||||
"prompt_eval_count": 1670,
|
||||
"eval_count": 51,
|
||||
"content_len": 0,
|
||||
"tool_call_count": 1,
|
||||
"history_chars_before_append": 3914
|
||||
},
|
||||
{
|
||||
"step": 4,
|
||||
"elapsed_s": 2.11,
|
||||
"prompt_eval_count": 1806,
|
||||
"eval_count": 86,
|
||||
"content_len": 311,
|
||||
"tool_call_count": 0,
|
||||
"history_chars_before_append": 4188
|
||||
}
|
||||
],
|
||||
"final": {
|
||||
"halt_reason": "no_tool_calls",
|
||||
"steps_used": 4,
|
||||
"tool_calls_total": 3,
|
||||
"wall_clock_s": 5.31,
|
||||
"final_message_count": 20,
|
||||
"final_history_chars": 4499
|
||||
},
|
||||
"task": "movies",
|
||||
"task_prompt": "Recommend 3 sci-fi movies NOT already in my sethflix library. Check your picks against check_sethflix before finalizing."
|
||||
}
|
||||
Reference in New Issue
Block a user