© 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. 30-TRANSCODER-HP
    7. 95950
    Prev
    Next
    INDEX
    Explanations

    Hatred, intolerance, prejudice

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éĵ
    -0.31
    isas
    -0.28
    客
    -0.27
    ä¸Ģä¸ĸ
    -0.26
    çĸ«
    -0.26
     SHR
    -0.26
    repid
    -0.25
    ä¾µ
    -0.24
    ãģĵãĤį
    -0.24
    åĻª
    -0.24
    POSITIVE LOGITS
     hatred
    0.34
    æŀģ端
    0.34
     intolerance
    0.33
     prejudice
    0.32
     fear
    0.31
     hate
    0.31
     tolerance
    0.30
     extreme
    0.30
     divisive
    0.29
     fanatic
    0.29
    Activations Density 0.893%

    No Known Activations