finding

active

finding:emotion-probe-persistence-correlation-of-0-214-in-cogito-v2-1-vs-0-099-for-random-vectors

Emotion probe persistence correlation of 0.214 in Cogito v2.1 vs 0.099 for random vectors

Quantifies emotion feature persistence above random baseline in Cogito across 240 multi-turn conversations

Source paper

extracted_from

Persistence and Introspection of Emotion Features

Scott Sauers · Imago · Janus · Antra Tessera

Neighborhood — ranked by edge-count

Claims (1)

claim

Emotion features in LLMs are genuinely more persistent than variance-matched random features, indicating stateful emotional encoding beyond autoregressive dynamics
supports
Central interpretive claim of the paper supported by multiple convergent analyses

Findings (1)

finding

Emotion probe persistence (token-0 to token-100 correlation) in Cogito v2.1 is 0.214, compared to 0.099 for random unit vectors in 7168D space.
restates
Quantitative measure of emotion feature persistence vs random baseline in Cogito

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.

Cogito emotion probe residual autocorrelation +0.077 above variance-matched controls (p=1.5e-27, 157/171 probes positive)finding0.884
Demonstrates that Cogito emotion probes are persistently active beyond what is explained by their variance alone
In Cogito v2.1, average residual persistence above variance-matched probes is +0.077 (p = 1.5e-27, 157 of 171 probes positive).finding0.842
Demonstrates emotion-specific persistence beyond variance effects in Cogito
SAE emotion subspace overlap correlates with variance-residualized persistence in Cogito: Spearman +0.413, p = 4.4e-196.finding0.830
Strong positive relationship between emotion alignment and SAE feature persistence in Cogito
SAE feature emotion subspace overlap correlates with persistence in Cogito: Spearman +0.413, p=4.4e-196finding0.820
Demonstrates that SAE features more aligned with the emotion subspace are more persistent in Cogito after variance control
Emotion probes are more persistent than variance-matched random probes, indicating emotion-specific persistence beyond autoregressive dynamics.claim0.810
Core empirical claim distinguishing emotion persistence from generic high-variance probe persistence
Negative correlation between self-evaluated emotion persistence and SAE feature activation variance explained: rho=-0.184, p=4.6e-09finding0.809
Shows self-evaluated emotionality is negatively confounded by variance, requiring variance control to reveal the true signal
Lower (more central) PCs of emotion feature activations are more persistent than higher-rank (noisier) PCs in both Kimi and Cogito, above variance-matched baselines.finding0.802
Supports that persistence is genuinely tied to emotion structure rather than measurement artifact
Lower (more central) emotion PCs are more persistent than higher (noisier) PCs in both Kimi and Cogitofinding0.790
Rules out that persistence is an artifact of probe construction, since noise dimensions are not similarly persistent

Restated by (1)

cosine ≥ 0.90

Other entities that say roughly the same thing. May be merge candidates or independent restatements across papers.

finding
Emotion probe persistence (token-0 to token-100 correlation) in Cogito v2.1 is 0.214, compared to 0.099 for random unit vectors in 7168D space.