© 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. 22-GEMMASCOPE-RES-131K
    7. 67783
    Prev
    Next
    INDEX
    Explanations

    selection and hiring processes

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

    The neuron is acting as a positional encoder—it fires more strongly on tokens appearing later in the text.

    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
    ͌
    -1.19
    znak
    -1.18
     levering
    -1.15
    허
    -1.14
     Aunque
    -1.14
     گیاه
    -1.13
    WHY
    -1.10
     каждо
    -1.07
    ಇ
    -1.06
    maillot
    -1.05
    POSITIVE LOGITS
     candidates
    1.30
    ּוֹ
    1.29
     selection
    1.28
    lendir
    1.27
     options
    1.27
     selecting
    1.20
    ִּ
    1.18
     zamiast
    1.16
     preference
    1.16
     hiring
    1.14
    Activations Density 0.017%

    No Known Activations