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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 25445
    Prev
    Next
    INDEX
    Explanations

    suspicious intentions and deaths

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

    The neuron activates strongly on numbers and numeric expressions (including decimals, ranges, and percentages) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ynthesis
    -0.91
    ophilic
    -0.90
    woven
    -0.89
    jungan
    -0.88
    kick
    -0.86
    lipop
    -0.85
     officially
    -0.84
    }]
    -0.84
    леми
    -0.83
     iga
    -0.82
    POSITIVE LOGITS
     suspicious
    1.99
     suspiciously
    1.73
     suspicion
    1.31
    Suspicious
    1.31
     mysterious
    1.21
     sospe
    1.17
     suspicions
    1.15
     unexplained
    1.14
     suggestive
    1.13
     ominous
    1.04
    Activations Density 0.062%

    No Known Activations