© 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. 43517
    Prev
    Next
    INDEX
    Explanations

    say single letters

    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
    y
    -0.35
    yw
    -0.29
    yt
    -0.28
    iona
    -0.27
     pres
    -0.27
    {'
    -0.27
    AGO
    -0.26
    ['
    -0.26
     running
    -0.26
    ppt
    -0.26
    POSITIVE LOGITS
    Ty
    0.27
    纵横
    0.27
    以åħ¶
    0.26
    åĴ¥
    0.26
    è¾£
    0.26
    åºŁ
    0.26
    丼æŀĹ
    0.26
    RB
    0.26
    æ³°å±±
    0.26
    æĵ¦æĭŃ
    0.26
    Activations Density 0.002%

    No Known Activations