© 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. 40-GEMMASCOPE-2-TRANSCODER-262K
    7. 246978
    Prev
    Next
    INDEX
    Explanations

    The TOP_POSITIVE_LOGITS are a mix of international-related words in different languages, foreign characters, and some English words that might be loosely related or simply co-occurring.Looking at the combined information:- `MAX_ACTIVATING_TOKENS` is overwhelmingly "International" (with variations in capitalization).- `TOKENS_AFTER_MAX_ACTIVATING_TOKEN` frequently shows "relations", "License", "Stock", "dispute", which are common collocations with "International".international relations or license

    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_40_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
    zten
    0.39
    Mobs
    0.39
    ucible
    0.36
    viles
    0.36
    inios
    0.36
     積
    0.36
     Traditions
    0.35
     ऑफ़
    0.35
     Chaotic
    0.35
    یه
    0.35
    POSITIVE LOGITS
    ﾛ
    0.43
     internacional
    0.41
     irrit
    0.41
     মাইল
    0.39
    领
    0.38
    の高い
    0.38
    羅
    0.38
    吝
    0.37
    Picklist
    0.36
     bilhões
    0.36
    Activations Density 0.001%

    No Known Activations