© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 73039
    Prev
    Next
    INDEX
    Explanations

    life improvement challenges

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

    This neuron is highly sensitive to section titles or headings—especially motivational/self-help headings—often rendered in title case and set apart (e.g., with colons or large font).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    וּ
    -1.46
     étages
    -1.40
    ᾿
    -1.33
     MECHANICAL
    -1.32
     taal
    -1.29
     vru
    -1.27
     saveurs
    -1.25
    的气息
    -1.25
     înc
    -1.24
     België
    -1.22
    POSITIVE LOGITS
     you
    2.73
     your
    2.44
     how
    1.74
     when
    1.74
     what
    1.66
     people
    1.66
     сахара
    1.65
     helps
    1.60
     says
    1.50
     starts
    1.46
    Activations Density 0.071%

    No Known Activations