© 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. 32-TRANSCODER-HP
    7. 42330
    Prev
    Next
    INDEX
    Explanations

    population and census

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     trait
    -0.29
    rete
    -0.28
    ret
    -0.26
    %A
    -0.26
    Else
    -0.26
    Sh
    -0.24
    trait
    -0.24
    _TA
    -0.24
    record
    -0.24
    Trap
    -0.24
    POSITIVE LOGITS
    ç«ĭæĸ¹
    0.28
    ç´¯äºĨ
    0.26
    -ton
    0.26
    TON
    0.26
    ton
    0.26
     cubic
    0.26
    ä¸ĩå¹³æĸ¹ç±³
    0.25
    æijĦ
    0.25
    .Tree
    0.25
    é¡¿
    0.24
    Activations Density 0.009%

    No Known Activations