© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 109285
    Prev
    Next
    INDEX
    Explanations

    incorrect rendering or display

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

    The neuron fires primarily on common natural‐language “function” words (e.g. but, with, since, I, get, some, before, right, etc.) in the post text rather than on code tokens.

    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
     sufficiently
    -1.04
     uncomplicated
    -1.03
     dibanding
    -1.00
    ‌.
    -0.99
     safely
    -0.98
     you
    -0.94
     hopefully
    -0.93
    踩
    -0.93
     comfortably
    -0.93
     inspired
    -0.92
    POSITIVE LOGITS
     instead
    1.36
     почему
    1.32
    になってしまう
    1.27
     weird
    1.26
     strange
    1.21
     strangely
    1.20
    instead
    1.20
     ujar
    1.20
     unintended
    1.19
    Instead
    1.15
    Activations Density 0.562%

    No Known Activations