Mortdecai
df5542f7d6
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>
12 lines
363 B
JSON
12 lines
363 B
JSON
{
|
|
"arm": "jax-native",
|
|
"error": "gemma package not importable \u2014 run in a JAX+gemma env. See module docstring.",
|
|
"final": {
|
|
"halt_reason": "env_missing",
|
|
"steps_used": 0,
|
|
"tool_calls_total": 0,
|
|
"wall_clock_s": 0
|
|
},
|
|
"task": "memory",
|
|
"task_prompt": "What do I have stored about home automation? If anything, summarize it briefly."
|
|
} |