© 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. 8293
    Prev
    Next
    INDEX
    Explanations

    hexadecimal strings

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

    The neuron flags tokens that look like long alphanumeric ID fragments (e.g. parts of commit hashes or object IDs).

    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
     to
    -2.48
     its
    -1.66
     among
    -1.45
     when
    -1.44
     which
    -1.41
     between
    -1.31
     アンド
    -1.30
     itself
    -1.25
     a
    -1.23
     with
    -1.22
    POSITIVE LOGITS
    
    1.58
     You
    1.45
    you
    1.45
    Sound
    1.32
    4
    1.31
     étan
    1.30
    This
    1.30
    Each
    1.28
    9
    1.28
    7
    1.26
    Activations Density 0.020%

    No Known Activations