© 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-3-4B
    3. 29-GEMMASCOPE-2-RES-16K
    4. 2625
    Prev
    Next
    INDEX
    Explanations

    rewriting and improving text

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-4b-pt/resid_post/layer_29_width_16k_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
     deflection
    0.67
     columnar
    0.64
     분포
    0.64
     assembly
    0.63
    삽
    0.62
     Filed
    0.61
    FieldSet
    0.60
    埋
    0.60
    到達
    0.59
    設定
    0.58
    POSITIVE LOGITS
     renovations
    2.86
     renovation
    2.85
     renovate
    2.80
     refurb
    2.80
     revamp
    2.78
     refurbishment
    2.77
     upgrade
    2.73
     upgrades
    2.72
     upgrading
    2.68
     renovating
    2.66
    Activations Density 0.715%

    No Known Activations