© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 102264
    Prev
    Next
    INDEX
    Explanations

    HTML DOCTYPE PUBLIC

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

    the term "PUBLIC" in DOCTYPE declarations and HTML document headers.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron fires on tokens that are capitalized or in all-caps—i.e. proper nouns, acronyms, or other words starting with an uppercase letter.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     you
    -1.92
     always
    -1.51
     they
    -1.46
     also
    -1.41
     a
    -1.36
    These
    -1.30
     
    -1.30
     []:
    -1.29
    Always
    -1.29
     these
    -1.28
    POSITIVE LOGITS
    íu
    1.38
    neté
    1.34
     with
    1.27
     그는
    1.24
    strategie
    1.23
    superuser
    1.22
    ResetPassword
    1.22
    Suara
    1.20
     and
    1.20
     読む
    1.20
    Activations Density 0.165%

    No Known Activations