© 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. 109113
    Prev
    Next
    INDEX
    Explanations

    academic lectures and seminars

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

    The neuron is detecting long runs of leading whitespace/indentation (i.e. large blank or spaced‐out segments at the start of lines).

    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
    还不
    -0.76
    Много
    -0.75
     Mujeres
    -0.74
    findung
    -0.73
    ――
    -0.73
    thold
    -0.72
    Lol
    -0.72
     Velas
    -0.72
    </>
    -0.71
    ──
    -0.70
    POSITIVE LOGITS
     seminar
    2.33
     seminars
    1.84
     collo
    1.63
     Seminar
    1.61
    seminar
    1.55
     talks
    1.41
    Seminar
    1.41
     talk
    1.40
     семина
    1.37
     lunch
    1.33
    Activations Density 0.073%

    No Known Activations