© 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. 108770
    Prev
    Next
    INDEX
    Explanations

    Ko

    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
    éĽı
    -0.27
    tain
    -0.26
    icine
    -0.25
    æ¼Ķèīº
    -0.24
    å½±
    -0.24
     exc
    -0.24
    æĿĢèıĮ
    -0.24
    xFE
    -0.24
    جتÙħاع
    -0.24
    pch
    -0.24
    POSITIVE LOGITS
    SP
    0.27
    uts
    0.25
    orr
    0.25
    (SP
    0.24
    (fi
    0.24
    sg
    0.24
    jug
    0.23
    åł¡
    0.23
    -su
    0.23
    _INLINE
    0.23
    Activations Density 0.288%

    No Known Activations