© 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. 13536
    Prev
    Next
    INDEX
    Explanations

    compound nouns or specific roles

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

    This neuron activates on tokens that are names or identifiers—essentially words written in title case or all caps (proper nouns or code identifiers).

    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
     estrateg
    -0.94
    rolex
    -0.93
    ubahan
    -0.93
    inspira
    -0.93
    imagination
    -0.92
     всички
    -0.91
    decap
    -0.90
    nivea
    -0.88
    angkut
    -0.88
    especial
    -0.88
    POSITIVE LOGITS
    ap
    0.96
     miami
    0.90
    CM
    0.90
    ma
    0.88
    p
    0.88
    M
    0.88
    r
    0.88
    E
    0.87
    uc
    0.87
    mam
    0.86
    Activations Density 0.021%

    No Known Activations