© 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. 30-TRANSCODER-HP
    7. 96123
    Prev
    Next
    INDEX
    Explanations

    Direction or position relative to something

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    -live
    -0.31
    æ´»
    -0.30
    -quote
    -0.28
    leting
    -0.28
    èĨ
    -0.26
    ativ
    -0.25
    aar
    -0.25
    caret
    -0.24
    ativos
    -0.24
    еÑģÑĤв
    -0.24
    POSITIVE LOGITS
     in
    0.35
    walker
    0.33
     whatever
    0.30
    à¹ĥà¸Ļ
    0.30
    åĩ«
    0.27
     whichever
    0.26
    -center
    0.26
     around
    0.25
    çIJ¥
    0.25
     somewhere
    0.25
    Activations Density 0.103%

    No Known Activations