question_name: {{ question }}

Exception details

{% for exception_message in exceptions %}
Exception: {{ exception_message.name }}
Key Value
Interview ID (index in results) {{ index }}
Question name (question_name) {{ question }}
Question type (question_type) {{ exception_message.question_type }}
Human-readable question {{ interview.survey._get_question_by_name(question).html( scenario = interview.scenario, agent = interview.agent, answers = exception_message.answers) }}
Scenario {{ interview.scenario.__repr__() }}
Agent {{ interview.agent.__repr__() }}
Model name {{ interview.model.model }}
Inference service {{ interview.model._inference_service_ }}
Model parameters {{ interview.model.__repr__() }}
User Prompt
{{ exception_message.rendered_prompts['user_prompt'] }}
System Prompt
{{ exception_message.rendered_prompts['system_prompt'] }}
Raw model response
{{ exception_message.raw_model_response }}
Generated token string (at {{ exception_message.key_sequence }}) in raw response
{{ exception_message.generated_token_string }}
Code to (likely) reproduce the error
{% if exception_message.exception.__class__.__name__ == 'QuestionAnswerValidationError' %}

Answer validation details

{% for field, (explanation, open_tag, close_tag, value) in exception_message.exception.to_html_dict().items() %} {% endfor %}
Field Value
{{ field }}: ({{ explanation }}) <{{open_tag}}> {{ value | escape }} <{{close_tag}}>
{% endif %}
Time: {{ exception_message.time }}
Traceback:
{{ exception_message.traceback }}
{% endfor %}