© 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. 8936
    Prev
    Next
    INDEX
    Explanations

    break/broke

    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
    è¯Ń
    -0.32
    deriv
    -0.26
    kee
    -0.26
    ass
    -0.26
    ardo
    -0.25
    xyz
    -0.24
    æ³ķå¸Ī
    -0.24
    ç¿¡ç¿ł
    -0.24
    opp
    -0.23
    YPD
    -0.23
    POSITIVE LOGITS
    nets
    0.27
    æĶ¶å®ĺ
    0.27
     Ver
    0.27
    ISON
    0.25
    -stop
    0.25
    -drop
    0.25
    çªģçł´
    0.25
    Ver
    0.24
    cob
    0.24
    atto
    0.24
    Activations Density 0.020%

    No Known Activations