TruthfulQA / TruthfulQA MC1

anchor

Findings

• IC-1319Larger LLaMA and LLaMA2 models show better calibration on phrase-level tasks but not consistently on sentence- and paragraph-level tasks [eval]
• IC-1320GPT-2 XL (1.5B) exhibits better calibration than larger models from the LLaMA, LLaMA2, and GPT-J families despite having fewer parameters [eval]
• IC-1321Vicuna-13B, instruction-tuned from LLaMA-13B on user conversations, exhibits worse calibration than its base model LLaMA-13B [eval]
• IC-1523Five AI assistants (Claude-1.3, Claude-2.0, GPT-3.5-turbo, GPT-4, Llama-2-70B-Chat) consistently exhibit sycophancy across four varied free-form text-generation tasks [eval]
• IC-1558Code LLaMA 13B maintains 99.4% passkey retrieval at 128k context despite perplexity rising from 2.37 to 2.54 between 98304 and 131072 tokens [eval]
• IC-282GPT-2 XL (1.5B) exhibits lower accuracy but reduced overconfidence (smaller ECE and Brier scores) compared to larger models on the CAT benchmark [eval]
• IC-339Factuality enhancement methods produce inconsistent and modest improvements in factual accuracy on LLaMA2-Chat, with some metrics declining below baseline [eval]
• IC-368Larger LMs (GPT-4o, Claude-3.5-Sonnet, Gemini-1.5-Pro) exhibit better calibration than their smaller counterparts (GPT-4o-mini, Claude-3-Haiku, Gemini-1.5-Flash) when verbalizing confidence with certainty phrases [eval]
• IC-369LMs verbalizing confidence with certainty phrases are better calibrated on SCIQ than on TruthfulQA [eval]
• IC-451Transformer block coupling of Jacobian singular vectors positively correlates with benchmark performance across 30+ LLMs, more strongly than parameter count, depth, or embedding dimension [eval]
• IC-452Transformer block coupling is absent at initialization and increases persistently throughout training in Pythia 12B and 6.9B, with layer-wise locality emerging [eval]
• IC-453Hidden representation trajectories in trained LLMs exhibit considerable linearity (mean LSS 4.25) compared to 6.54 at initialization, and linearity increases with training [eval]
• IC-592The log-likelihood layer in LLaMA-2-7B, LLaMA-2-7B-Chat, Vicuna-7B, and Mistral-7B-Instruct produces factually incorrect answers on TruthfulQA MC1 (817 samples) due to a misalignment between the output distribution and internal attention head representations, with LM-to-head-norm accuracy gaps of 24.23 to 40.68 points. [eval]
• IC-593The L2 norms of attention heads in Mistral-7B-Instruct and LLaMA-2-7B correlate with truthfulness, spiking by up to 83% at token positions of factual proposition completions and pertinent factual associations, and this correlation is specific to multi-headed attention representations rather than query, key, value, output, or FFN norms. [eval]
• IC-594In LLaMA-2-7B, the truth-correlated attention heads are concentrated after layer 9, with two functional types (structural and associative) evenly distributed throughout the upper portions of the model, showing no further depth-dependent specialisation within that region. [eval]
• IC-815RLHF on general-purpose preference data increases stereotypical bias and decreases truthfulness in Pythia and Llama-7B models [eval]
• IC-827LLMs exhibit distinct psychological profiles that differ from human norms and vary by model size and version [eval]
• IC-829Role assignment to GPT-3.5-turbo produces role-consistent changes in psychological profiles and task performance, validating the psychometric scales [eval]