© 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. 22-TRANSCODER-HP
    7. 98639
    Prev
    Next
    INDEX
    Explanations

    Conditionals, questions, and answers

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lements
    -0.28
    鸳
    -0.27
    adows
    -0.27
     rog
    -0.27
    quier
    -0.26
    è½®
    -0.24
    amble
    -0.23
    cá
    -0.23
     Imm
    -0.23
     Concord
    -0.23
    POSITIVE LOGITS
    plib
    0.25
    é¼Ļ
    0.25
     Universal
    0.24
     telesc
    0.24
    /util
    0.24
    ä¹łæĥ¯äºĨ
    0.24
    çĬĢ
    0.24
    æľīäºĨ
    0.24
     utilise
    0.24
    volent
    0.23
    Activations Density 0.002%

    No Known Activations