© 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. 142118
    Prev
    Next
    INDEX
    Explanations

    necessity/desire

    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
    æĺ¯ä¸Ģ件
    -0.30
     sonst
    -0.27
    AT
    -0.27
     Anyone
    -0.27
     atlas
    -0.26
     anyone
    -0.26
     ai
    -0.25
     Then
    -0.25
     recon
    -0.24
     Recon
    -0.23
    POSITIVE LOGITS
    款
    0.29
    èĩ³ä¸Ĭ
    0.27
    unger
    0.26
     thicker
    0.26
    çĶŁäºİ
    0.26
    èĸª
    0.26
    åij¨å²ģ
    0.26
    诣
    0.24
     Territory
    0.23
    thic
    0.23
    Activations Density 17.753%

    No Known Activations