© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 82423
    Prev
    Next
    INDEX
    Explanations

    deception and delusion

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on first‐person self–referential words and negations (e.g. I, me, myself, self, not, pretend), i.e. introspective/self-focused language.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    たた
    -0.71
    problems
    -0.70
     Gerald
    -0.69
     razu
    -0.68
    izar
    -0.68
    верху
    -0.67
    tingham
    -0.67
    imed
    -0.67
     combineReducers
    -0.66
    Length
    -0.66
    POSITIVE LOGITS
     fooling
    2.81
     kidding
    2.77
     delu
    2.45
     deceive
    2.44
     deceiving
    2.38
     deluded
    2.27
     illusion
    2.09
     delusion
    2.06
     deceived
    1.98
     fool
    1.96
    Activations Density 0.030%

    No Known Activations