© 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. 107843
    Prev
    Next
    INDEX
    Explanations

    Random encoding/code/languages

    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
    wards
    -0.32
    æĭ§
    -0.28
    rons
    -0.28
    饧
    -0.27
    ism
    -0.26
    ness
    -0.26
    ding
    -0.26
    Wh
    -0.26
    esimal
    -0.26
    tical
    -0.26
    POSITIVE LOGITS
    åħ¨å¸Ĥ
    0.31
     improv
    0.30
     terra
    0.28
    çħ§
    0.27
    家åħ·
    0.26
     floors
    0.26
    infra
    0.25
    家ãģ®
    0.25
     presets
    0.25
    fur
    0.25
    Activations Density 0.979%

    No Known Activations