© 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. 22-GEMMASCOPE-RES-131K
    4. 10952
    Prev
    Next
    INDEX
    Explanations

    continue reading or more information

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

    The neuron detects runs of periods—especially ellipses or long strings of dot characters.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     in
    -1.35
     only
    -1.35
     when
    -1.35
    While
    -1.34
     so
    -1.32
     or
    -1.28
     While
    -1.27
     because
    -1.24
    !(
    -1.21
     if
    -1.20
    POSITIVE LOGITS
    рил
    1.30
     produkcji
    1.26
     výrob
    1.13
     cóc
    1.13
    つい
    1.11
    躓
    1.11
    夘
    1.10
    欤
    1.10
     utilizado
    1.09
    だら
    1.09
    Activations Density 0.001%

    No Known Activations