© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 76816
    Prev
    Next
    INDEX
    Explanations

    official firearm theft

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

    The neuron activates on formal, factual or technical expository prose (e.g. encyclopedic, academic, patent‐style) rather than on casual or narrative 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_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
     género
    -0.74
    koku
    -0.73
    Sä
    -0.71
     газ
    -0.70
    UV
    -0.69
    mación
    -0.68
     manuscript
    -0.68
    컨
    -0.67
    bericht
    -0.66
    HEA
    -0.66
    POSITIVE LOGITS
     issued
    5.47
    issued
    4.56
     issuing
    4.34
     issuance
    4.19
    Issued
    4.03
     Issued
    4.03
     issue
    3.61
     ISSUED
    3.25
    issue
    3.25
    Issue
    3.11
    Activations Density 0.069%

    No Known Activations