© 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. 89388
    Prev
    Next
    INDEX
    Explanations

    Code and/or copyright notices

    np_max-act · gemini-2.0-flash
    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
     mut
    -0.27
    æİ¥è¿ĩ
    -0.27
     verbs
    -0.27
    åĮ¹éħį
    -0.26
     matching
    -0.26
    etz
    -0.25
     Mut
    -0.25
     pir
    -0.25
    ually
    -0.24
    matching
    -0.24
    POSITIVE LOGITS
     shade
    0.28
    ble
    0.28
    è¿ĺæľīä¸Ģ个
    0.25
     beforeSend
    0.25
    éĤ£ä¹Ī容æĺĵ
    0.25
    лем
    0.24
    éĤ£ä¹Īç®Ģåįķ
    0.24
     essays
    0.24
    æĻĸ
    0.24
     cap
    0.24
    Activations Density 78.494%

    No Known Activations