© 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. 27-TRANSCODER-HP
    7. 58219
    Prev
    Next
    INDEX
    Explanations

    removing

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    udos
    -0.31
    /INFO
    -0.26
    orm
    -0.25
    å¨ĺ家
    -0.25
     conformity
    -0.25
    è´¹çİĩ
    -0.24
     mon
    -0.23
    æĮĤåľ¨
    -0.23
    forg
    -0.23
    orta
    -0.23
    POSITIVE LOGITS
    åİ»éĻ¤
    0.55
     removal
    0.49
     remove
    0.48
     removing
    0.47
     Removal
    0.47
     Removing
    0.46
     removed
    0.46
     Remove
    0.45
    remove
    0.44
     removes
    0.42
    Activations Density 0.001%

    No Known Activations