© 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. 108188
    Prev
    Next
    INDEX
    Explanations

    code or data structures

    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
    Moder
    -0.30
     Ul
    -0.29
     explosions
    -0.26
    flows
    -0.26
    ieten
    -0.26
    Ul
    -0.25
     Moderate
    -0.25
     modulation
    -0.24
    æĤ¬
    -0.24
     Moderator
    -0.24
    POSITIVE LOGITS
    asar
    0.27
    åĨį说
    0.26
    è§īå¾Ĺ
    0.25
     carried
    0.25
    çł»
    0.25
    .getOwnProperty
    0.25
    ä¹ĭåĬĽ
    0.25
    entes
    0.25
    ocl
    0.25
     Affero
    0.24
    Activations Density 0.015%

    No Known Activations