© 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. 34-GEMMASCOPE-RES-131K
    7. 62384
    Prev
    Next
    INDEX
    Explanations

    questions and phrases like "such as"

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

    This neuron fires strongly on numeric and hyphenated model or specification tokens (e.g. the “BLL-16” in “BLL-16 semi-automatic pyrolysis plant”), so it’s detecting product model numbers or similar technical spec identifiers.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    أك
    -0.93
    Mari
    -0.81
    樹
    -0.79
    ʋ
    -0.79
    Παραπομπές
    -0.77
    ierten
    -0.76
     ВА
    -0.76
    Ἀ
    -0.75
    ători
    -0.75
    Harry
    -0.75
    POSITIVE LOGITS
     adopts
    0.94
    Adopt
    0.84
    Sinon
    0.80
     baratas
    0.79
     rationality
    0.79
     מן
    0.78
     تغییر
    0.77
    Kelsey
    0.77
    Adop
    0.77
     maar
    0.76
    Activations Density 0.024%

    No Known Activations