© 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. Qwen3-4B
    5. 25-TRANSCODER-HP
    7. 107475
    Prev
    Next
    INDEX
    Explanations

    numbers

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ÑĨев
    -0.29
    crawl
    -0.27
    elow
    -0.27
    ancode
    -0.27
    _WM
    -0.26
    bish
    -0.26
    çīĪæĿĥæīĢæľī
    -0.26
    MOOTH
    -0.25
    廨
    -0.25
    Vien
    -0.25
    POSITIVE LOGITS
     two
    0.33
     three
    0.32
    两
    0.31
     six
    0.30
    äºĶ个
    0.30
    ä¸ī个
    0.29
    æĶ¶
    0.28
    èµİ
    0.28
    åIJĮç±»
    0.28
     four
    0.28
    Activations Density 0.890%

    No Known Activations