© 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. 88877
    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
    æĹ¢
    -0.26
     banged
    -0.25
    aina
    -0.25
    åį³
    -0.25
    itates
    -0.24
     Guar
    -0.24
    常è§ģ
    -0.24
    ÙĬات
    -0.24
    ücü
    -0.23
    stown
    -0.23
    POSITIVE LOGITS
    á»ģ
    0.29
    ä¸ĩ象
    0.27
    æĹłæĥħ
    0.27
    è´¨éĩı管çIJĨ
    0.27
    æ´Ĺ
    0.26
     warm
    0.25
    TIM
    0.25
    éªĦ
    0.25
    å¹²äºĭ
    0.25
     prep
    0.25
    Activations Density 35.617%

    No Known Activations

    This feature has no known activations.