© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 56-GEMMASCOPE-2-TRANSCODER-262K
    7. 20398
    Prev
    Next
    INDEX
    Explanations

    respond to queries

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_56_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
    குரு
    0.37
    뷁
    0.37
    Universität
    0.36
    Wiki
    0.35
    }$(
    0.34
    BTW
    0.34
    ڊ
    0.34
    SIMPLEX
    0.33
    subreddit
    0.33
    ၊
    0.33
    POSITIVE LOGITS
     shakespeare
    1.30
     nigeria
    1.30
     russia
    1.29
     california
    1.27
     england
    1.27
     delhi
    1.25
     dna
    1.23
     mexico
    1.23
     chicago
    1.23
     toyota
    1.21
    Activations Density 0.205%

    No Known Activations