© 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. 22-TRANSCODER-HP
    7. 130286
    Prev
    Next
    INDEX
    Explanations

    comma

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ablo
    -0.27
    COR
    -0.26
     Immediate
    -0.26
     "+"
    -0.26
    alo
    -0.25
    ä½İä½į
    -0.25
     Carlo
    -0.25
     mod
    -0.24
    æĭ¿äºĨ
    -0.24
    åŃĺåľ¨éĹ®é¢ĺ
    -0.24
    POSITIVE LOGITS
    ERING
    0.26
    itez
    0.25
    kn
    0.25
    åĮĸè¿Ľç¨ĭ
    0.25
    足足
    0.25
    career
    0.24
    ç»Ĩ
    0.24
    èīĩ
    0.24
    :variables
    0.23
    GetString
    0.23
    Activations Density 0.004%

    No Known Activations