© 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. 20-TRANSCODER-HP
    7. 107811
    Prev
    Next
    INDEX
    Explanations

    through

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ded
    -0.28
    èİ«åIJįåħ¶å¦Ļ
    -0.28
     Assigned
    -0.27
    çĶ¨äºİ
    -0.26
     stays
    -0.26
    æĶĺ
    -0.26
     handful
    -0.26
    umbs
    -0.25
    loy
    -0.24
     stupidity
    -0.24
    POSITIVE LOGITS
    çİ°åľº
    0.29
     Stone
    0.27
    çı¾åł´
    0.26
    æĽ´è¦ģ
    0.26
    ê°±
    0.25
    Stone
    0.25
    ossa
    0.25
     stone
    0.24
    ObjectContext
    0.24
     after
    0.23
    Activations Density 2.443%

    No Known Activations