© 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. 12-TRANSCODER-HP
    7. 23644
    Prev
    Next
    INDEX
    Explanations

    will

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ayer
    -0.28
    rgb
    -0.28
    维度
    -0.27
    then
    -0.26
     drinkers
    -0.25
     then
    -0.25
    CDC
    -0.24
    à¤ĺ
    -0.24
     dimensional
    -0.24
    buckets
    -0.24
    POSITIVE LOGITS
    umes
    0.29
    æīĵ好
    0.28
    illes
    0.26
    åıªä¼ļ
    0.25
    好人
    0.25
    ISR
    0.24
    WithType
    0.24
    llib
    0.24
    好äºĭ
    0.24
     minHeight
    0.24
    Activations Density 0.001%

    No Known Activations