© 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. 34-GEMMASCOPE-RES-131K
    7. 65969
    Prev
    Next
    INDEX
    Explanations

    filter and its contexts

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

    The neuron predominantly fires on occurrences of the word “filter” (in any casing or morphological form) and closely related terms.

    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
    s
    -1.14
     entram
    -1.08
     venit
    -0.97
    之所以
    -0.96
     ノート
    -0.95
    Contenu
    -0.95
     formalin
    -0.93
    鵟
    -0.93
     heißen
    -0.93
     vetorial
    -0.93
    POSITIVE LOGITS
     Filter
    1.80
     filter
    1.59
    filter
    1.58
     filters
    1.36
     Filters
    1.34
    Filter
    1.28
     filtro
    1.28
     Fil
    1.24
    过滤
    1.20
    filtr
    1.18
    Activations Density 0.023%

    No Known Activations