© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 90338
    Prev
    Next
    INDEX
    Explanations

    natural and textured hair

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

    The neuron strongly activates on occurrences of the word “hair” (and closely connected heading phrases like “Hair Meet-up”), i.e. it detects when the text is about hair.

    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
     stud
    -0.90
    hill
    -0.85
    宰
    -0.82
    清掃
    -0.77
    pran
    -0.76
     servo
    -0.73
    ὄ
    -0.73
     rife
    -0.72
     devo
    -0.72
    kle
    -0.72
    POSITIVE LOGITS
     edges
    0.91
     coils
    0.91
     twa
    0.90
    Transition
    0.90
     crecimiento
    0.89
    🏾
    0.87
    Curl
    0.86
     Curly
    0.86
     Afro
    0.86
     transitioning
    0.86
    Activations Density 0.014%

    No Known Activations