© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 48287
    Prev
    Next
    INDEX
    Explanations

    Deadpool, complexion, fleeting

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

    The neuron preferentially fires on longer, multi-syllable content words (e.g. “complexion,” “mediocrity,” “ineffable,” “fleeting”), i.e. unusually long descriptive terms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    }$
    -3.17
    –
    -3.00
     ibadah
    -2.81
     of
    -2.75
    </sub>
    -2.73
    он
    -2.69
    しばらく
    -2.58
    It
    -2.52
    々は
    -2.48
    —
    -2.44
    POSITIVE LOGITS
    section
    3.00
     
    2.64
     I
    2.53
     prominence
    2.39
     wichtiger
    2.36
     portrayal
    2.33
    出て
    2.30
    珰
    2.27
     tierras
    2.22
    彀
    2.20
    Activations Density 0.035%

    No Known Activations