© 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
    2. Gemma-3-27B-IT
    3. 38-GEMMASCOPE-2-TRANSCODER-262K
    4. 4352
    Prev
    Next
    INDEX
    Explanations

    I cannot provide a concise explanation based on the provided lists. The `MAX_ACTIVATING_TOKENS` list contains only periods (.), which does not reveal a specific semantic pattern or token sequence on its own. This prevents me from identifying what the neuron is detecting or predicting, as is required by the instructions to find patterns in words or tokens

    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_38_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
     आतंक
    0.41
     переме
    0.40
    Serial
    0.39
    깎
    0.39
    कभी
    0.36
    టీఎం
    0.36
     kurulu
    0.35
    fZ
    0.35
     ral
    0.35
    규모
    0.35
    POSITIVE LOGITS
    ains
    0.38
     يونيو
    0.37
     Milan
    0.36
     pecado
    0.36
    δι
    0.36
     holistic
    0.35
     Lavender
    0.35
     ibid
    0.34
    steelblue
    0.34
     Cap
    0.34
    Activations Density 0.001%

    No Known Activations