© 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. 18-TRANSCODER-HP
    7. 89726
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    潦
    -0.28
    Liked
    -0.25
    失æķĪ
    -0.24
    åīįèĢħ
    -0.24
    Pont
    -0.24
     practiced
    -0.24
    çļĦåŁºç¡Ģ
    -0.24
    ãĥ¯ãĤ¤
    -0.24
    sters
    -0.23
    %=
    -0.23
    POSITIVE LOGITS
    exterity
    0.28
     labour
    0.26
    letic
    0.26
     wrench
    0.25
    åľºä¸Ĭ
    0.24
    ibel
    0.24
    /ajax
    0.24
    tür
    0.24
    titre
    0.24
     turnover
    0.23
    Activations Density 0.923%

    No Known Activations

    This feature has no known activations.