© 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. 34-GEMMASCOPE-RES-131K
    7. 79005
    Prev
    Next
    INDEX
    Explanations

    not responding or working

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

    The neuron strongly activates on technical/log‐style or domain‐specific content words (e.g. error‐report and measurement terms) in system messages and code/log excerpts.

    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
    without
    -1.42
     fără
    -1.27
     senza
    -1.16
     zonder
    -1.09
     ohne
    -1.08
     without
    -1.03
    毫不
    -1.03
     totul
    -1.02
    sätzlich
    -1.02
    provided
    -1.02
    POSITIVE LOGITS
     nor
    2.19
     or
    2.13
     properly
    2.09
     adequately
    1.91
     adequate
    1.91
     anymore
    1.81
     because
    1.71
     sufficient
    1.62
     despite
    1.59
     enough
    1.51
    Activations Density 0.171%

    No Known Activations