© 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. 98834
    Prev
    Next
    INDEX
    Explanations

    .g

    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
    divider
    -0.30
    gil
    -0.28
    (proc
    -0.27
    ilar
    -0.26
     poc
    -0.25
     stil
    -0.25
     Attachment
    -0.24
    åŁºè¾ħ
    -0.24
    ICES
    -0.23
     ------------------------------------------------------------------------------------------------
    -0.23
    POSITIVE LOGITS
    æĦıä¹īä¸Ĭ
    0.28
    ãģĤãģŁãĤĬ
    0.26
    ombie
    0.25
    LOB
    0.25
    æĦıä¹īä¸ĬçļĦ
    0.24
    åħ´èµ·
    0.23
    åºĬåŀ«
    0.23
    å¾Ĵ
    0.23
    ory
    0.22
    ä»ĸä¼ļ
    0.22
    Activations Density 0.021%

    No Known Activations