© 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. 97569
    Prev
    Next
    INDEX
    Explanations

    for

    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
    åİ»
    -0.28
     Closet
    -0.27
    SL
    -0.26
     placement
    -0.26
    -placement
    -0.25
    pective
    -0.25
     Placement
    -0.24
    FLT
    -0.24
    .GO
    -0.24
     Intersection
    -0.23
    POSITIVE LOGITS
    uku
    0.28
    ninger
    0.28
    大åĨĽ
    0.26
     Gra
    0.25
     utilized
    0.25
    æ¶Īè´¹åįĩ级
    0.24
    æĤ²åī§
    0.24
    èħIJè´¥
    0.24
    æĹİ
    0.24
    utz
    0.24
    Activations Density 0.006%

    No Known Activations