© 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. 33-TRANSCODER-HP
    7. 53933
    Prev
    Next
    INDEX
    Explanations

    Game development/3D modeling

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĢģ
    -0.25
     TÃłi
    -0.24
    éĴĪ
    -0.24
    è´¨åľ°
    -0.24
     manifest
    -0.24
    vo
    -0.24
    elyn
    -0.24
    Vo
    -0.23
     Salt
    -0.23
    ocop
    -0.23
    POSITIVE LOGITS
     ik
    0.38
     Ik
    0.35
    ik
    0.34
    Ik
    0.33
     IK
    0.31
    ç½ij讯
    0.31
    çŃī级
    0.30
     hierarchy
    0.28
    fk
    0.28
    emos
    0.28
    Activations Density 0.165%

    No Known Activations