© 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. 35-TRANSCODER-HP
    7. 39418
    Prev
    Next
    INDEX
    Explanations

    Corrupted / nonsensical text

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     de
    -0.26
    SocketAddress
    -0.26
    .mods
    -0.24
    纣
    -0.24
    çļĦæĥħ
    -0.24
    icz
    -0.23
    åĬ¨æīĭ
    -0.23
     clo
    -0.23
    è¥Ħ
    -0.23
     Centers
    -0.23
    POSITIVE LOGITS
    orde
    0.28
    ogne
    0.28
    ordinate
    0.27
     Caught
    0.26
    ÑĢ
    0.26
    isse
    0.26
    or
    0.25
    orig
    0.25
    all
    0.25
    æ´Ĺå¹²åĩĢ
    0.25
    Activations Density 5.966%

    No Known Activations