© 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. 111668
    Prev
    Next
    INDEX
    Explanations

    largely

    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
    really
    -0.28
     rencontres
    -0.27
    chal
    -0.26
    Really
    -0.26
    straight
    -0.26
     keer
    -0.26
    little
    -0.25
    çķ¥å¾®
    -0.24
    ç¨įå¾®
    -0.24
    ologically
    -0.24
    POSITIVE LOGITS
    itud
    0.28
    iman
    0.27
     irrelevant
    0.25
    _TC
    0.25
     similar
    0.25
    æĬ¢
    0.24
    itude
    0.24
    æ¤Ĵ
    0.24
    pat
    0.23
    -dom
    0.23
    Activations Density 0.103%

    No Known Activations