© 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. 149561
    Prev
    Next
    INDEX
    Explanations

    Opinions and choices

    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
    &W
    -0.28
    åŁİå¸ĤçļĦ
    -0.27
    æŶéĹ´åĨħ
    -0.26
    çİ·
    -0.26
     Panel
    -0.25
    éĿ¢æĿ¿
    -0.25
    &E
    -0.25
    æŶéĹ´åĴĮ
    -0.24
     Keys
    -0.24
    sWith
    -0.24
    POSITIVE LOGITS
     Ari
    0.29
    agog
    0.27
    urpose
    0.27
    ä¸ºçĽ®çļĦ
    0.26
     before
    0.26
    以åīį
    0.25
     when
    0.25
    hots
    0.25
    为æŃ¢
    0.25
    为äºĨ
    0.25
    Activations Density 0.123%

    No Known Activations