© 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
    3. 31-GEMMASCOPE-2-RES-262K
    4. 66884
    Prev
    Next
    INDEX
    Explanations

    classic bracelet

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

    closing sequences*Self-correction:* The original prompt asked for a concise explanation (3 to 20 words) that captures what the neuron detects or predicts by finding patterns in lists.The MAX_ACTIVATING_TOKENS include punctuation `)`, `"`, `.`, and words like `equipment`, `Now`.The TOKENS_AFTER_MAX_ACTIVATING_TOKEN include `**(`, `Good`, `muscle`, `has`, `**(` , `Just`, `,`.The TOP_POSITIVE_LOGITS are foreign words related to selection (`เลือก`, `Escol`, `પસંદ`), showing (`Mostrar`), introducing (`kenalkan`), parameters (`ParamNum`), and specific items/concepts like `equipment`, `LockButton` (lock), `Tampa`.The TOP_ACTIVATING_TEXTS show examples involving:1. `equipment` and its access.2. `Now` followed by instructions like "don't you fidget", "Close your eyes", "good boy/girl". instructions after "Now"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    える
    0.51
     istor
    0.43
     İran
    0.43
     abe
    0.43
     avions
    0.43
     Niels
    0.43
    豫
    0.42
     器
    0.42
     convi
    0.41
     Bihar
    0.41
    POSITIVE LOGITS
     પસંદ
    0.42
    лон
    0.42
    Escol
    0.42
    Clique
    0.42
    FName
    0.42
    することが
    0.41
     منذ
    0.41
    ParamNum
    0.41
    LockButton
    0.40
    जामा
    0.40
    Activations Density 0.002%

    No Known Activations