© 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. 31-TRANSCODER-HP
    7. 18726
    Prev
    Next
    INDEX
    Explanations

    Name prefixes/parts in legal context

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MD
    -0.31
     gag
    -0.26
    USA
    -0.26
     Mach
    -0.26
    avo
    -0.26
    亮
    -0.25
    енÑĮ
    -0.25
    æıĸ
    -0.25
    athy
    -0.24
    Ñĥв
    -0.24
    POSITIVE LOGITS
    æľīä¸Ģå®ļ
    0.27
    shal
    0.26
    ä¸ºæľŁ
    0.26
     behalf
    0.25
     getIntent
    0.25
    IEL
    0.25
    µľ
    0.25
    è¿ģ移
    0.24
     manip
    0.24
     Manip
    0.24
    Activations Density 0.478%

    No Known Activations