© 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. 34-GEMMASCOPE-RES-131K
    4. 75475
    Prev
    Next
    INDEX
    Explanations

    code and file names

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

    The neuron spikes on standalone punctuation and symbol tokens (e.g. semicolons, slashes, dots) commonly used as separators in code and URLs.

    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
     sous
    -0.74
     pivot
    -0.74
    xtext
    -0.68
    扭
    -0.66
     BENE
    -0.66
     up
    -0.66
     Empirical
    -0.66
    setPos
    -0.65
    omé
    -0.65
    iar
    -0.65
    POSITIVE LOGITS
     Händ
    0.74
    jetivo
    0.68
    nourriture
    0.62
    legging
    0.60
    CLEAN
    0.59
    манд
    0.58
    interesting
    0.58
     trabajar
    0.58
    MSB
    0.58
     Work
    0.57
    Activations Density 0.062%

    No Known Activations