© 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. 12221
    Prev
    Next
    INDEX
    Explanations

    colon followed by code or value

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

    closing angle brackets or colons in HTML/XML tags and code formatting contexts.

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

    The neuron is triggered by colon characters (":"), especially when used to separate a label from its value.

    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
    ，
    -1.59
     his
    -1.52
     one
    -1.46
    ?
    -1.45
     ホテル
    -1.45
     primeira
    -1.39
     mais
    -1.34
    ONE
    -1.34
    、
    -1.32
    首先
    -1.30
    POSITIVE LOGITS
    ifrance
    1.45
    甭
    1.45
     nup
    1.39
    となっています
    1.38
    ឵
    1.35
    いため
    1.34
    putan
    1.34
     mien
    1.32
    になっています
    1.32
     rud
    1.32
    Activations Density 0.008%

    No Known Activations