© 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. 24-TRANSCODER-HP
    7. 150919
    Prev
    Next
    INDEX
    Explanations

    (

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åĪĻ
    -0.33
    ãĢĤ
    -0.32
    `.
    -0.29
    |.
    -0.29
    .".
    -0.28
    .";
    -0.28
    ."
    -0.27
    æīį
    -0.27
    ".
    -0.27
    ”.
    -0.27
    POSITIVE LOGITS
     which
    0.30
    åħ¶ä¸ŃåĮħæĭ¬
    0.28
    è¡ħ
    0.27
    èĤ´
    0.25
     wives
    0.25
    -inv
    0.25
    omy
    0.24
    èĢĮä¸Ķè¿ĺ
    0.24
    redict
    0.24
    —which
    0.24
    Activations Density 0.118%

    No Known Activations