© 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. 31651
    Prev
    Next
    INDEX
    Explanations

    app endings

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

    The neuron fires on tokens that are names of software projects, libraries, modules, or other code-related identifiers.

    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
     and
    -2.45
     this
    -2.06
     what
    -1.95
     one
    -1.74
     any
    -1.68
     an
    -1.65
     some
    -1.64
     supremely
    -1.60
     another
    -1.58
     from
    -1.54
    POSITIVE LOGITS
     unending
    1.66
     exuber
    1.51
    often
    1.51
     dismal
    1.50
     REALLY
    1.46
     devast
    1.45
    mainly
    1.45
     overarching
    1.45
     acclaimed
    1.44
     horrid
    1.44
    Activations Density 0.063%

    No Known Activations