© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 30778
    Prev
    Next
    INDEX
    Explanations

    contrast or exception introducers

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

    The neuron fires on language pitching commercial or financial opportunity—words and phrases about making money, market size, “huge” demand, “lots” of profit, etc.

    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
    .”
    -1.58
     though
    -1.57
    ."
    -1.35
    <eos>
    -1.29
    》。
    -1.20
     for
    -1.18
    though
    -1.17
     in
    -1.17
     хотя
    -1.14
    ”.
    -1.11
    POSITIVE LOGITS
     BUT
    3.56
    However
    3.55
    BUT
    3.16
     However
    3.13
     But
    3.08
     however
    2.86
     แต่
    2.67
    But
    2.63
     Однако
    2.30
     però
    2.28
    Activations Density 0.302%

    No Known Activations