Contribution to the handbook
Set aside · Assessed 2026-10-28
This is a narrow study-method detail. The evidence identifies LLaMA-2 7B among the fine-tuned models but establishes no material governance-design, risk, implementation, or testing implication for the handbook.
Does this challenge the goal?
None identified
The question asks for a study-specific model identifier and does not contest the exploration’s goal or identify a missing governance requirement.
The supplied evidence names models used in one study but does not connect a particular model choice to governance design.
Why this category?
- The model identifier is peripheral to the handbook’s system-design goal; no concrete design or testing implication is established.
This assessment concerns the supplied question and evidence, not the general value of comparing models.
This is an AI assessment of contribution to the goal, separate from whether the question has an answer.
New evidence is being assessed. Any answer below is provisional; parts whose evidence changed or was withdrawn are withheld.
Give the specific LLaMA model identifier and reported model description.
What we can say so far
The researchers identified two LLaMA models for fine-tuning: LLaMA-2 7B and LLaMA-3 8B. They describe LLaMA-2 and LLaMA-3 as open-weight models trained on proprietary data. The reported fine-tuning used collected policy-preference data.
What remains open
No material gap is recorded within this part’s current scope.
Conditions and limits
- This reports the authors’ method and model description in the supplied preprint text; it is not independent verification.
- The fine-tuning is described in the context of the study’s collected policy-preference data.
Evidence and related ideas
Supporting evidence, qualifications and context play different roles in this answer.
supports
Six fine-tuned models The authors fine-tune six existing language models on the volunteer data: Llama-2 7B, LLaMA-3 8B, ChatGPT 3.5 Turbo, Mistral 7B, Falcon 7B, and Gemma 7B.
Lists LLaMA-2 7B and LLaMA-3 8B among the models fine-tuned on volunteer data.
Large Language Models (LLMs) as Agents for Augmented Democracy
Read full source: Large Language Models (LLMs) as Agents for Augmented Democracy (opens in a new tab) ↗Read source discussion: Large Language Models (LLMs) as Agents for Augmented Democracy →
supports
Models in the comparison The comparison includes six language models and BERT SQuAD, a simpler benchmark. The models differ in what is known about their training: GPT 3.5 Turbo has proprietary architecture and data; LLaMA-2, LLaMA-3, Mistral, and Gemma are open-weight models trained on proprietary data; Falcon is open-weight and trained on public data; and BERT is a foundation model trained on public data.
Describes LLaMA-2 and LLaMA-3 as open-weight models trained on proprietary data.
Large Language Models (LLMs) as Agents for Augmented Democracy
Read full source: Large Language Models (LLMs) as Agents for Augmented Democracy (opens in a new tab) ↗Read source discussion: Large Language Models (LLMs) as Agents for Augmented Democracy →
context
Predictions improved some sample matches In the Type I B test, the authors fine-tuned six models on the same preference data: GPT-3.5 Turbo, LLaMA-2 7B, Falcon 7B, Mistral 7B, LLaMA-3 8B and Gemma 7B. They tested initial random samples of 5%, 25% and 50%, adding predictions for 20% of the people who remained outside each sample. For the 5% sample, R² rose from about 30% to about 75%; for the 25% sample, the gain was about 7 to 10 percentage points. The supplied text gives no separate numerical result for the 50% sample.
Reiterates both LLaMA model identifiers in its account of a performance test; the test results are not needed to identify the models.