© 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. 10-GEMMASCOPE-RES-131K
    7. 100202
    Prev
    Next
    INDEX
    Explanations

    elusive and words ending in sive

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

    The neuron is tuned to spot adjectives ending in the “-ive” (or “-sive”) suffix.

    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
    “
    -3.42
    —
    -2.61
     gehörte
    -2.55
    因為
    -2.48
    ができます
    -2.44
    素晴らしい
    -2.42
     the
    -2.42
     veicolo
    -2.39
     potere
    -2.38
    \
    -2.31
    POSITIVE LOGITS
    sive
    2.89
    麈
    2.70
     onCreateView
    2.56
    窠
    2.53
    ing
    2.48
    炱
    2.48
     padx
    2.42
     ).
    2.33
    èvent
    2.33
    戾
    2.28
    Activations Density 0.008%

    No Known Activations