© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 6715
    Prev
    Next
    INDEX
    Explanations

    extracellular domains and invasive cancer

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

    code and technical syntax patterns, particularly in markup languages and programming contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on tokens that belong to document‐level headers, metadata fields, or technical labels (e.g. section titles, file‐ or contract‐IDs, code comments, XML attribute names).

    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
    Rp
    -0.73
    oniec
    -0.70
    đe
    -0.69
     désigne
    -0.69
    Peny
    -0.68
    黑白
    -0.68
    Reward
    -0.66
     Kinn
    -0.66
    tyy
    -0.66
    itoneal
    -0.66
    POSITIVE LOGITS
    勲
    0.66
    ICLE
    0.66
    risa
    0.65
    ullo
    0.64
    BLEND
    0.62
    刚才
    0.61
    Aquatic
    0.61
    #######
    0.60
    lite
    0.59
    NGS
    0.59
    Activations Density 0.086%

    No Known Activations