© 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. 85071
    Prev
    Next
    INDEX
    Explanations

    words ending in ic or ish

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

    The neuron responds strongly to words using the “-ic” suffix (e.g. “Pythonic,” “un-American,” etc.).

    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
     sebag
    -1.30
     for
    -1.22
     quello
    -1.20
    ària
    -1.13
     în
    -1.12
     нього
    -1.11
     Generally
    -1.10
     Like
    -1.10
    IALOG
    -1.10
     héroe
    -1.09
    POSITIVE LOGITS
    3
    1.20
     blanda
    1.17
     formul
    1.17
    และ
    1.16
     people
    1.15
    шим
    1.15
    typical
    1.10
    while
    1.09
     должна
    1.09
     Ferr
    1.09
    Activations Density 0.090%

    No Known Activations