© 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. 19194
    Prev
    Next
    INDEX
    Explanations

    Internet protocols and networking

    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
     pierced
    -0.26
     Cru
    -0.26
    eyer
    -0.26
    anford
    -0.25
     Dudley
    -0.25
     dis
    -0.25
    pac
    -0.25
     striped
    -0.25
     bevor
    -0.24
    æ¶Ķ
    -0.24
    POSITIVE LOGITS
    ipop
    0.28
    è°Ĵ
    0.27
     CIT
    0.26
    ellipsis
    0.26
    enus
    0.26
    Bas
    0.26
     Fet
    0.26
     Tec
    0.26
     RT
    0.25
    flen
    0.25
    Activations Density 0.942%

    No Known Activations