© 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. 98681
    Prev
    Next
    INDEX
    Explanations

    top

    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
    TEE
    -0.29
    een
    -0.28
    tor
    -0.25
    åıijåĬ¨
    -0.24
     doi
    -0.24
    دب
    -0.24
     Loud
    -0.24
    eba
    -0.24
    _listing
    -0.24
    åĮ¹
    -0.24
    POSITIVE LOGITS
    -notch
    0.36
    pling
    0.32
    =top
    0.30
    篷
    0.29
    /top
    0.29
    ,top
    0.28
     notch
    0.27
    -top
    0.27
    ographical
    0.27
    顶éĥ¨
    0.27
    Activations Density 0.066%

    No Known Activations