© 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-3-27B-IT
    3. 20-GEMMASCOPE-2-TRANSCODER-262K
    4. 208538
    Prev
    Next
    INDEX
    Explanations

    **TOP_ACTIVATING_TEXTS**: * "...accurate modeling..." * "...segment should be reachable through marketing channels." * "...broader democratic aspirations..."Combining these observations:- `accurate` followed by `modeling`- `reachable` followed by `through` (suggesting something is reachable *through* some medium)- `democratic` followed by `aspirations`The `MAX_ACTIVATING_TOKENS` are `accurate`, `reachable`, `democratic`.The `TOKENS_AFTER_MAX_ACTIVATING_TOKEN` are `modeling`, `through`, `aspirations`.The pattern is descriptive words followed by related nouns or prepositions indicating purpose, means, or outcome."accurate modeling""reachable through X""democratic aspirations"The common theme is evaluative or defining qualities followed by what they lead to or are associated with.qualities followed by outcomes

    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_20_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
    ಟ್
    0.44
    ি
    0.44
    следования
    0.43
     estánd
    0.42
    ACTIVITY
    0.42
     affirmat
    0.41
    っ
    0.41
    蛄
    0.40
     ফর
    0.39
     joyería
    0.39
    POSITIVE LOGITS
    ષા
    0.46
     Buna
    0.45
     serve
    0.44
     lingu
    0.43
    層
    0.43
     ظرف
    0.42
     धु
    0.42
    äre
    0.41
     fumes
    0.41
    烬
    0.41
    Activations Density 0.000%

    No Known Activations