© 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. 25-TRANSCODER-HP
    7. 108746
    Prev
    Next
    INDEX
    Explanations

    Code and foreign languages

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    DV
    -0.28
    pert
    -0.26
    DIC
    -0.26
     pert
    -0.26
    å®ŀåĬ¡
    -0.25
    åıĹçIJĨ
    -0.25
    brook
    -0.25
    åĩºæīĭ
    -0.25
    runtime
    -0.24
    æ¹ĸæ³Ĭ
    -0.24
    POSITIVE LOGITS
    .ua
    0.27
    âĤ¬
    0.26
     &)
    0.26
    å¾Ĺèµ·
    0.25
     Eu
    0.25
    缮çļĦ
    0.24
    对ä¸ŃåĽ½
    0.24
     quoi
    0.24
    çIJĨçĶ±
    0.24
    oted
    0.23
    Activations Density 0.103%

    No Known Activations