© 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-3-27B-IT
    5. 16-GEMMASCOPE-2-TRANSCODER-262K
    7. 213797
    Prev
    Next
    INDEX
    Explanations

    The provided data seems to indicate the neuron is primarily focused on identifying French words, particularly those related to actions or descriptors. Looking at `MAX_ACTIVATING_TOKENS`, we see `par` (part of pre

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_16_width_262k_l0_small_affine
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     تلیفون
    0.48
     drunkenness
    0.48
    \%)
    0.46
     기준으로
    0.46
     повіт
    0.46
    вут
    0.46
    }^{+
    0.45
    苚
    0.45
     electrification
    0.45
     reheat
    0.45
    POSITIVE LOGITS
     Say
    0.55
     Detail
    0.51
    p
    0.48
     Details
    0.48
     Stabil
    0.47
     Ensure
    0.47
     Question
    0.46
     Resource
    0.46
    i
    0.46
    Say
    0.45
    Activations Density 0.000%

    No Known Activations