© 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. 94138
    Prev
    Next
    INDEX
    Explanations

    code, copyright, license

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

    This neuron fires on document‐structuring tokens—section headers, titles, metadata keys, and other markers that signal block or heading boundaries rather than ordinary running text.

    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
     આ
    -0.80
     Vaters
    -0.78
    dbh
    -0.76
     Orc
    -0.73
     Apple
    -0.72
     orc
    -0.72
    LOAT
    -0.72
    Orc
    -0.71
    Potion
    -0.71
     Easy
    -0.71
    POSITIVE LOGITS
     Corpus
    1.00
     Syndicate
    0.93
    Kuva
    0.92
     kuva
    0.89
     polarity
    0.88
     wf
    0.83
     syndicate
    0.83
    charpe
    0.82
    frame
    0.80
     Forma
    0.80
    Activations Density 0.012%

    No Known Activations