© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 40028
    Prev
    Next
    INDEX
    Explanations

    connective tissue

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

    words related to connections or connectivity, particularly the term "connexon" and variations of "connect" in scientific and technical contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on occurrences of the root “connec-/connex-,” i.e. words built from the “connect” or “connex” stem.

    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
    硖
    -2.66
    弱点
    -2.48
    瑒
    -2.36
    同情
    -2.34
     pazi
    -2.22
    戗
    -2.20
     Ausblick
    -2.19
    卧槽
    -2.16
    驚き
    -2.16
    万一
    -2.16
    POSITIVE LOGITS
    ”
    2.69
     "
    2.66
    »
    2.53
    ,,
    2.50
    ╽
    2.42
    圌
    2.33
    «
    2.31
     così
    2.30
    𝄆
    2.27
    <b>
    2.22
    Activations Density 0.003%

    No Known Activations