© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 38516
    Prev
    Next
    INDEX
    Explanations

    physical objects, medical conditions, places

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

    technical or specialized nouns related to different domains like construction materials, medical conditions, and place names.

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

    The neuron activates on longer, multi-syllable content words—especially domain-specific nouns and acronyms.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    compound and proper nouns, especially co-/counter‑prefixed terms and similar closed compounds.

    oai_token-act-pair · gpt-5
    Triggered by @jyhe0408

    words that are part of proper nouns or brand names, particularly place names and company/product names.

    oai_token-act-pair · claude-4-5-sonnet
    Triggered 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
    Negative Logits
    䜣
    -2.84
    which
    -2.75
    of
    -2.72
    𞤢
    -2.58
    晅
    -2.48
    aktery
    -2.39
     verschiedener
    -2.39
    ids
    -2.39
    März
    -2.33
    𞤮
    -2.33
    POSITIVE LOGITS
     However
    3.36
     become
    2.86
     “
    2.48
     Aprend
    2.45
     According
    2.42
     But
    2.28
     additionally
    2.25
     nena
    2.06
     furthermore
    2.06
     willingly
    2.03
    Activations Density 0.005%

    No Known Activations