© 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. 103610
    Prev
    Next
    INDEX
    Explanations

    ankle and Achilles

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

    anatomical references to ankles and related parts of the lower leg.

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

    The neuron strongly activates on occurrences of the anatomical term “ankle” (including its appearance in compound and morphological variants).

    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
    umm
    -0.75
     CallOverrides
    -0.72
    меты
    -0.71
    centaje
    -0.71
     nœ
    -0.71
     الحي
    -0.68
    neros
    -0.68
    ographic
    -0.66
    MARTIN
    -0.66
     perbedaan
    -0.66
    POSITIVE LOGITS
     ankle
    2.50
     ankles
    2.41
     Ankle
    2.05
    ankle
    1.87
     legs
    1.13
     leg
    1.11
     Ank
    1.09
     feet
    1.07
     bracelet
    0.98
     ank
    0.96
    Activations Density 0.057%

    No Known Activations