© 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. 18-TRANSCODER-HP
    7. 88971
    Prev
    Next
    INDEX
    Explanations

    code and data

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lean
    -0.28
    artial
    -0.27
    stores
    -0.26
    æĿİåħĪçĶŁ
    -0.26
    éķ¿å¾Ĺ
    -0.26
    ilst
    -0.25
    avia
    -0.25
     NotÃŃcias
    -0.25
    zeÅĦ
    -0.25
    派人
    -0.25
    POSITIVE LOGITS
    ?key
    0.25
    æĢª
    0.24
    æĴĤ
    0.24
     !_
    0.23
    )?$
    0.23
     unify
    0.23
    ERING
    0.23
    >?
    0.23
    éĺĸ
    0.23
    Callable
    0.23
    Activations Density 4.593%

    No Known Activations