© 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. 10-GEMMASCOPE-RES-131K
    7. 21098
    Prev
    Next
    INDEX
    Explanations

    parts and components

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

    The neuron fires strongly on words naming parts of a firearm (e.g. barrel, stock, bipod, hand guard, grip).

    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
     about
    -1.42
     manifiesta
    -1.35
    čený
    -1.31
    asantry
    -1.30
     entweder
    -1.30
     alleviation
    -1.28
     exemplu
    -1.28
     Entwurf
    -1.23
     fortæller
    -1.22
    hlás
    -1.20
    POSITIVE LOGITS
     fece
    1.53
    皤
    1.49
     csin
    1.49
     cobr
    1.47
     tomar
    1.42
    橚
    1.41
     dado
    1.40
    AYLOR
    1.36
     como
    1.36
    
    1.34
    Activations Density 0.035%

    No Known Activations