© 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. 88876
    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
    proxy
    -0.26
     discharge
    -0.25
    por
    -0.25
    uation
    -0.24
    vak
    -0.24
    åį³ä¾¿æĺ¯
    -0.23
     issuance
    -0.23
    ç¬Ĩ
    -0.23
    ип
    -0.23
    quence
    -0.22
    POSITIVE LOGITS
     nid
    0.26
    substr
    0.26
     SharedModule
    0.24
    _substr
    0.24
    æĻļ
    0.23
    å°ģ
    0.23
    éĵį
    0.23
    æĹ¥æĻļ
    0.23
    itta
    0.23
     limit
    0.23
    Activations Density 15.320%

    No Known Activations

    This feature has no known activations.