© 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. 27-TRANSCODER-HP
    7. 9029
    Prev
    Next
    INDEX
    Explanations

    formal vocabulary

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     dept
    -0.29
    .googleapis
    -0.27
    gte
    -0.26
    ivity
    -0.26
    éĢĢåĽŀ
    -0.26
    fäh
    -0.26
    pong
    -0.25
    ä»°
    -0.25
     vite
    -0.24
     chạy
    -0.24
    POSITIVE LOGITS
     axiom
    0.41
    足
    0.37
     inconsistent
    0.32
     ax
    0.32
     formal
    0.31
    表述
    0.30
     consistent
    0.30
     consistency
    0.29
    consistent
    0.29
     formulas
    0.28
    Activations Density 0.021%

    No Known Activations