© 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. 34-GEMMASCOPE-RES-131K
    7. 63874
    Prev
    Next
    INDEX
    Explanations

    especially considering

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

    The neuron fires strongly on the main content‐bearing words of a post—especially topic keywords like “training,” “disciplined,” “war,” “Vietnam,” “organizing,” etc.

    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
     благодаря
    -1.52
     dzięki
    -1.32
     grâce
    -1.32
     согласно
    -1.22
     due
    -1.19
     according
    -1.18
     līdz
    -1.12
     thanks
    -1.11
    according
    -1.09
     entweder
    -1.05
    POSITIVE LOGITS
     especially
    2.58
    Especially
    2.13
     Especially
    2.09
     столь
    2.09
    しかも
    1.98
    especially
    1.89
     considering
    1.80
     knowing
    1.80
     when
    1.80
     surtout
    1.71
    Activations Density 0.071%

    No Known Activations