© 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. 22-GEMMASCOPE-RES-131K
    7. 26540
    Prev
    Next
    INDEX
    Explanations

    mathematical digit operations

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

    The neuron is triggered by phrasing that asks for a particular place‐value digit in a number—i.e. ordinal modifiers like “ten,” “hundreds,” “thousands,” etc., immediately followed by “digit.”

    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
     such
    -1.52
    such
    -1.09
     appunto
    -1.02
    Such
    -1.02
    ntilde
    -0.94
     preferably
    -0.94
     основы
    -0.94
     sufficiently
    -0.93
     Such
    -0.92
    newblock
    -0.91
    POSITIVE LOGITS
     chacune
    1.04
    edakan
    1.02
    觀眾
    0.99
     demokra
    0.98
    явление
    0.98
     деца
    0.96
     peluche
    0.91
    //"
    0.90
    Ų
    0.89
    چون
    0.88
    Activations Density 0.071%

    No Known Activations