© 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. 75269
    Prev
    Next
    INDEX
    Explanations

    additionally, acrescent, ditionally

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

    The neuron is a “tech‐Q&A” detector: it fires on code/markup tokens (e.g. `<li>`, `</li>`), programming‐ or site‐specific jargon and acronyms (e.g. “response,” “Server,” “Firefox,” “GSB”), and other boilerplate words you’d see in StackOverflow‐style questions and answers.

    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
     so
    -1.18
     although
    -1.09
     therefore
    -0.94
     both
    -0.91
     hence
    -0.89
     because
    -0.86
     only
    -0.85
     though
    -0.84
     despite
    -0.82
    de
    -0.81
    POSITIVE LOGITS
    țiunea
    1.04
     inclut
    1.03
     samtidig
    1.03
     LEAVE
    0.98
    Ajouter
    0.98
     acresc
    0.97
    ditional
    0.96
     acrescent
    0.96
     additionally
    0.93
    ditionally
    0.93
    Activations Density 0.304%

    No Known Activations