© 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. 89035
    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
    uego
    -0.27
     reassuring
    -0.25
     vaccinated
    -0.25
     Sultan
    -0.25
     Verified
    -0.24
     decay
    -0.24
    puts
    -0.24
    纸ä¸Ĭ
    -0.23
    滩
    -0.23
    _decay
    -0.23
    POSITIVE LOGITS
    ä¸ĵ线
    0.30
     Hook
    0.26
     settlement
    0.26
     hook
    0.24
     bÃŃ
    0.24
    HOOK
    0.24
    æľĢé«ĺçļĦ
    0.24
    å±ĢéĿ¢
    0.24
    ç¾¹
    0.24
    [js
    0.24
    Activations Density 2.595%

    No Known Activations

    This feature has no known activations.