© 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. 18-TRANSCODER-HP
    7. 88966
    Prev
    Next
    INDEX
    Explanations

    prevention and protection

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    amt
    -0.28
    .getInstance
    -0.27
    mis
    -0.27
    åı¯éĢļè¿ĩ
    -0.27
    Mis
    -0.26
     mis
    -0.26
    ward
    -0.25
    å½ĵä¹ĭ
    -0.25
    çľ¼æ³ª
    -0.24
    第äºĮæī¹
    -0.24
    POSITIVE LOGITS
    europ
    0.26
    oten
    0.26
     condom
    0.25
     clubhouse
    0.25
     randomness
    0.25
    hoff
    0.24
    åIJįåĪĹ
    0.24
    便
    0.24
     innovations
    0.23
     stacking
    0.23
    Activations Density 14.839%

    No Known Activations