© 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. 108317
    Prev
    Next
    INDEX
    Explanations

    Boilerplate/code/legal text

    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
    èª
    -0.28
    sky
    -0.28
    说åĪ°
    -0.27
     motivated
    -0.26
    lik
    -0.26
    æİ©çĽĸ
    -0.26
    #${
    -0.26
    -to
    -0.26
     setting
    -0.25
    ä¸Ńå¿ĥ
    -0.25
    POSITIVE LOGITS
    pare
    0.30
    unei
    0.26
    uggage
    0.26
    unos
    0.26
    former
    0.25
    箸
    0.25
    ortex
    0.25
    èħĬ
    0.24
     raw
    0.24
    anship
    0.24
    Activations Density 0.891%

    No Known Activations