fix: eliminate duplicate checkpoint entries and JSON-unsafe coercion

batch_runner: completed_prompts_set is already fully populated by the
time the aggregation loop runs (incremental updates happen at result
collection time), so the subsequent extend() call re-added every
completed prompt index a second time. Removed the redundant variable
and extend, and write sorted(completed_prompts_set) directly to the
final checkpoint instead.

model_tools: _coerce_number returned Python float('inf')/float('nan')
for inf/nan strings rather than the original string. json.dumps raises
ValueError for these values, so any tool call where the model emitted
"inf" or "nan" for a numeric parameter would crash at serialization.
Changed the guard to return the original string, matching the
function's documented "returns original string on failure" contract.
This commit is contained in:
debthemelon 2026-04-24 13:33:56 +01:00 committed by Teknium
parent db9d6375fb
commit dbdefa43c8
2 changed files with 4 additions and 8 deletions

View file

@ -951,13 +951,9 @@ class BatchRunner:
root_logger.setLevel(original_level)
# Aggregate all batch statistics and update checkpoint
all_completed_prompts = list(completed_prompts_set)
total_reasoning_stats = {"total_assistant_turns": 0, "turns_with_reasoning": 0, "turns_without_reasoning": 0}
for batch_result in results:
# Add newly completed prompts
all_completed_prompts.extend(batch_result.get("completed_prompts", []))
# Aggregate tool stats
for tool_name, stats in batch_result.get("tool_stats", {}).items():
if tool_name not in total_tool_stats:
@ -977,7 +973,7 @@ class BatchRunner:
# Save final checkpoint (best-effort; incremental writes already happened)
try:
checkpoint_data["completed_prompts"] = all_completed_prompts
checkpoint_data["completed_prompts"] = sorted(completed_prompts_set)
self._save_checkpoint(checkpoint_data, lock=checkpoint_lock)
except Exception as ckpt_err:
print(f"⚠️ Warning: Failed to save final checkpoint: {ckpt_err}")