© 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. 34-GEMMASCOPE-RES-131K
    7. 78231
    Prev
    Next
    INDEX
    Explanations

    code comments and licenses

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

    This neuron fires on uppercase technical tokens—especially acronyms and protocol/header identifiers (e.g. HMAC, SignedHeaders, Signature).

    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
     portraits
    -0.74
    jstor
    -0.70
    国民
    -0.68
     fleuve
    -0.68
     Kashi
    -0.67
    เรื่อง
    -0.66
    uage
    -0.65
    ]").
    -0.65
    keits
    -0.65
    said
    -0.64
    POSITIVE LOGITS
     nul
    0.75
    êne
    0.69
     Seoul
    0.69
    DataObject
    0.68
     هند
    0.67
    Seoul
    0.67
    sterone
    0.66
    馏
    0.65
    anten
    0.65
    ես
    0.65
    Activations Density 0.083%

    No Known Activations