© 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. 41500
    Prev
    Next
    INDEX
    Explanations

    lim

    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
    cin
    -0.30
    asti
    -0.28
    ITH
    -0.27
    orch
    -0.27
     cin
    -0.27
    bsp
    -0.26
    åħ¬æ°ij
    -0.26
    èĬ«
    -0.26
     Cin
    -0.25
    abstract
    -0.25
    POSITIVE LOGITS
    åĴ¯
    0.31
    åĬª
    0.27
    airo
    0.27
    ette
    0.26
    åħĶ
    0.26
    ocomplete
    0.26
    ç³ł
    0.26
    uts
    0.26
    头
    0.26
     sage
    0.26
    Activations Density 0.002%

    No Known Activations