© 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. 34-GEMMASCOPE-RES-131K
    4. 90081
    Prev
    Next
    INDEX
    Explanations

    python imports and declarations

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

    The neuron fires strongly on low-frequency, content-specific subword tokens (e.g. module or command names, proper nouns, or other rare identifiers).

    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
     halve
    -0.77
     antibio
    -0.75
    ढ़
    -0.71
    身上的
    -0.68
     Söder
    -0.68
    萦
    -0.67
    kenn
    -0.66
     murales
    -0.66
     multime
    -0.66
     ofc
    -0.66
    POSITIVE LOGITS
     rehabilitation
    0.68
     gene
    0.68
     bizony
    0.68
    :[],
    0.66
    askin
    0.66
    Consultation
    0.63
    habilitation
    0.63
    اشتی
    0.63
     blo
    0.63
     heißt
    0.62
    Activations Density 0.053%

    No Known Activations