© 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. 97636
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    nement
    -0.26
    osphate
    -0.25
    hot
    -0.24
     nomine
    -0.24
    ponce
    -0.24
     mond
    -0.24
    vrir
    -0.23
    hon
    -0.23
    verture
    -0.23
    aturity
    -0.23
    POSITIVE LOGITS
    ªĮ
    0.27
     dec
    0.27
    ren
    0.26
    IJèĹı
    0.25
    NAMESPACE
    0.25
    dbl
    0.25
    åĨ³å®ļ
    0.25
    .sess
    0.24
    ess
    0.24
    æ¸Ŀ
    0.24
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.