© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 91609
    Prev
    Next
    INDEX
    Explanations

    impression and impulse

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

    This neuron detects words containing the root “impress” (e.g. impression, impressed, impressionable).

    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
    N
    -1.31
    L
    -1.15
     It
    -1.14
    ھے
    -1.12
     recurring
    -1.08
    C
    -1.08
    让我
    -1.07
     pièce
    -1.06
    D
    -1.06
    必定
    -1.05
    POSITIVE LOGITS
     impresion
    1.24
     Imp
    1.23
    Imp
    1.21
     ciała
    1.21
    IMP
    1.17
     gepub
    1.08
     IMP
    1.04
    YMS
    1.02
    able
    1.02
    ਜ
    1.02
    Activations Density 0.039%

    No Known Activations