© 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. 33693
    Prev
    Next
    INDEX
    Explanations

    conjunctions and separators

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

    The neuron strongly activates on forward‐slash punctuation tokens (the “/” separators often used in lists or compound terms).

    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
    aneity
    -1.12
    ourced
    -1.11
     to
    -1.10
     ()
    -1.05
     whiteColor
    -1.03
    borderSide
    -1.02
    などを
    -1.00
     amerikanischen
    -1.00
     --->
    -0.97
    が
    -0.97
    POSITIVE LOGITS
    INCLUDING
    1.33
     queijo
    1.11
    🕔
    1.08
    Pozri
    1.07
     including
    1.03
     PLAINT
    1.02
    Herkunft
    1.00
    INCLUDES
    1.00
    nios
    0.99
    כול
    0.98
    Activations Density 0.027%

    No Known Activations