Post-training is key to eliciting introspective awareness

Base pretrained models show high false positive rates and achieve no net task performance on concept injection detection; post-training essential for introspection.

Source paper

extracted_from

Emergent Introspective Awareness in Large Language Models

(2026) · Lindsey, Jack

Neighborhood — ranked by edge-count

Claims (1)

claim

Post-training influences introspective capability expression
supports
Different post-training strategies substantially influence introspection task performance; 'helpful-only' variants show higher false positives but some achieve strong net performance.

Communities (4)

community

Mechanistic interpretability & model evaluation
members_of
Spans attention head decomposition, benchmark awareness, and genomic pathogenicity prediction via neural models.
Mechanistic introspection in language models
members_of
Empirical investigation of how LMs access and report internal states across layers, using concept injection and thought detection on Claude models.
LLM functional introspective awareness
members_of
Empirical probing of language models' ability to detect and report their own internal concept representations
Post-training emergence of model introspection
members_of
How instruction tuning and RLHF elicit latent introspective capabilities in language models beyond base pretraining.

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

Post-training is key to eliciting strong introspective awareness; base pretrained models do not show above-chance detectionclaim0.928
Finding that base models have high false positives and no net positive performance.
Post-training strategies can strongly influence performance on introspective tasksclaim0.871
Assertion about the role of post-training in eliciting introspection.
What are the mechanistic bases of introspective awareness in LLMs?question0.806
Secondary question; paper demonstrates introspection but explicitly avoids pinning down specific mechanistic explanation, noting mechanisms could be shallow and specialized.
Post-training elicits rather than installs personas; the persona direction is already laid down in pretrainingclaim0.805
Authors' interpretive endorsement of PSM view, backed by transfer experiments
Post-training steers models toward a particular region of persona space but only loosely tethers them to it, motivating work on training and steering strategies that more deeply anchor models to a coherent personaclaim0.803
Central interpretive claim and motivation for future work
Post-Trainingconcept0.800
The phase after pre-training where models are further tuned with techniques like DPO; the period where the studied behavior emerged.
Introspective signals appear in middle layers but are suppressed by later post-training-shaped layers.finding0.789
Mechanistic finding by Lindsey (2026) explaining how contemplative prompt may work: enables mid-layer introspection to reach output.
Contemplative training can lead to enhanced compassion, social connectedness, and ethical sensibility particularly when practices incorporate moral reflectionsclaim0.784
Empirical generalization from contemplative neuroscience supporting the viability of Contemplative AI approach