© 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. 19-TRANSCODER-HP
    7. 8826
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     thereof
    -0.31
    icum
    -0.27
    åı¶åŃIJ
    -0.25
    ä¸Ģæľµ
    -0.24
    |%
    -0.24
    yster
    -0.24
    _GEN
    -0.23
    女人
    -0.23
    çĻ¾èĬ±
    -0.23
     Walters
    -0.23
    POSITIVE LOGITS
    åħ°
    0.29
    raison
    0.28
    ses
    0.26
    eries
    0.25
    evenodd
    0.25
     situated
    0.25
    è¦ģåľ¨
    0.25
    EEDED
    0.24
     rigid
    0.24
    sed
    0.24
    Activations Density 0.001%

    No Known Activations