© 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. 35-TRANSCODER-HP
    7. 39430
    Prev
    Next
    INDEX
    Explanations

    admissions, confessions, acknowledgements

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Ack
    -1.14
    ack
    -1.13
     ack
    -1.13
     admission
    -1.10
    ACK
    -1.09
     admit
    -1.07
    Ack
    -1.05
     admitting
    -1.05
     ACK
    -1.02
     acknowledge
    -1.00
    POSITIVE LOGITS
     freely
    0.36
     effortlessly
    0.32
     bigotry
    0.30
     boycott
    0.30
     outraged
    0.30
     angered
    0.30
     distra
    0.29
     scorn
    0.29
     ironically
    0.29
     readily
    0.29
    Activations Density 11.791%

    No Known Activations