© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 67455
    Prev
    Next
    INDEX
    Explanations

    sequence of mathematical expressions and code

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

    This neuron spotlights code tokens that represent object member accesses or method/property calls indicated by a leading “.” (dot).

    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
     empêcher
    -1.00
     lèvres
    -0.96
     factoring
    -0.90
     enliven
    -0.90
     VIDEOT
    -0.90
    雳
    -0.87
    lillah
    -0.86
    ᭄
    -0.85
     ។
    -0.84
    સ
    -0.83
    POSITIVE LOGITS
     kabát
    1.10
    With
    1.02
     maroca
    1.00
    When
    0.93
     luxus
    0.92
     没
    0.90
    ʮ
    0.90
     to
    0.90
    ........
    0.90
     vulnerables
    0.90
    Activations Density 0.011%

    No Known Activations