© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 119504
    Prev
    Next
    INDEX
    Explanations

    working code

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

    possessive constructions or phrases indicating ownership or belonging.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires most strongly on content‐heavy, topic‐specific terms (especially proper nouns or technical product names) rather than on common function words.

    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
     우리
    -1.46
     itſelf
    -1.38
     douleurs
    -1.34
     impecable
    -1.30
    俢
    -1.30
     ergon
    -1.28
    荑
    -1.28
     korban
    -1.27
     jät
    -1.27
    höhe
    -1.25
    POSITIVE LOGITS
     same
    1.94
     el
    1.44
     infamous
    1.41
     o
    1.41
     C
    1.38
     ve
    1.33
     most
    1.33
     u
    1.32
     identical
    1.31
     l
    1.30
    Activations Density 0.080%

    No Known Activations