© 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. 34-GEMMASCOPE-RES-131K
    7. 23684
    Prev
    Next
    INDEX
    Explanations

    license headers and identifiers

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

    This neuron activates on document–formatting and structure tokens (e.g. “admonition,” “id,” “head,” “children,” “list-style,” etc.), i.e. markup directives and metadata labels.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    technical documentation and code comments, particularly license headers and metadata.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    keywords that denote document structure or metadata in technical text and markup formats.

    oai_token-act-pair · gpt-5Triggered 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
    ORIA
    -0.80
    Gweler
    -0.74
     Hermione
    -0.71
    inigungs
    -0.70
     MSS
    -0.70
    ⁎
    -0.68
     nig
    -0.67
     scuola
    -0.67
    ENTOS
    -0.66
    瑗
    -0.66
    POSITIVE LOGITS
     أص
    0.65
     gallows
    0.64
    ftest
    0.64
    Роз
    0.63
    決定
    0.62
    chieben
    0.62
    ضي
    0.61
    Dil
    0.61
    Formato
    0.60
     Perrin
    0.60
    Activations Density 0.036%

    No Known Activations