© 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. 107928
    Prev
    Next
    INDEX
    Explanations

    as

    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
    .inspect
    -0.28
     goto
    -0.26
    é¢Ĭ
    -0.26
    åĩºå¤Ħ
    -0.26
    èķĥ
    -0.25
    culture
    -0.25
     possessing
    -0.25
     Zah
    -0.24
     Locate
    -0.24
    ohana
    -0.24
    POSITIVE LOGITS
    @Web
    0.28
    สา
    0.27
    ucker
    0.26
    åĽ¢ç»ĵ
    0.26
     Cor
    0.26
    æīĭä¸ŃçļĦ
    0.26
     Cam
    0.25
    åĨ°
    0.25
     ice
    0.25
     Ca
    0.25
    Activations Density 0.903%

    No Known Activations