© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 477
    Prev
    Next
    INDEX
    Explanations

    The neuron responds to longer, formal descriptive adjectives (often multi-syllabic) that qualify or frame scientific or technical statements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Or
    0.34
     Normalmente
    0.32
     Insgesamt
    0.32
     besonderen
    0.31
     Eine
    0.31
    Eine
    0.31
     Only
    0.30
    `
    0.30
     Các
    0.30
    E
    0.30
    POSITIVE LOGITS
     failings
    0.41
     frustrations
    0.41
     nefarious
    0.41
     nerdy
    0.40
     bureaucratic
    0.40
     carnage
    0.40
     populace
    0.39
     semblance
    0.39
     societal
    0.39
     hassles
    0.39
    Activations Density 2.278%

    No Known Activations