© 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. 150278
    Prev
    Next
    INDEX
    Explanations

    tough

    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
    eton
    -0.25
    ience
    -0.25
     Basel
    -0.24
     Partition
    -0.24
    istant
    -0.24
    achel
    -0.23
    ointed
    -0.23
    INGTON
    -0.23
    询éĹ®
    -0.23
     Câmara
    -0.23
    POSITIVE LOGITS
    speed
    0.28
     resp
    0.26
    ä¸ĢæĬ¹
    0.25
    å¤ł
    0.25
     {}:
    0.25
    ä¸Ģæ³¢
    0.24
    GU
    0.24
    DER
    0.23
    kat
    0.23
     regeneration
    0.23
    Activations Density 2.468%

    No Known Activations