© 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. 10-GEMMASCOPE-RES-131K
    7. 11258
    Prev
    Next
    INDEX
    Explanations

    investigation and investigating

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

    The neuron detects occurrences of “investigat-” word stems (e.g. investigate, investigating, investigation, investigative).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    '
    -3.16
    </h2>
    -2.83
    </td>
    -2.73
    </h1>
    -2.41
    </
    -2.34
    (
    -2.30
    </b>
    -2.28
    ">
    -2.17
    €™
    -2.05
    [
    -2.03
    POSITIVE LOGITS
    ፧
    1.91
    しているので
    1.91
    Результа
    1.86
    蔟
    1.82
     Při
    1.82
    貽
    1.80
    な感じ
    1.80
    为他
    1.74
     nových
    1.72
    するので
    1.71
    Activations Density 0.072%

    No Known Activations