© 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. 14-TRANSCODER-HP
    7. 40587
    Prev
    Next
    INDEX
    Explanations

    are

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_14.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.14.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    nde
    -0.26
    éĢıè¿ĩ
    -0.25
    rette
    -0.24
    çı°
    -0.24
    tight
    -0.24
    ç²±
    -0.24
    GBK
    -0.24
    liner
    -0.23
    arcy
    -0.23
    çļĦæľĢä½³
    -0.23
    POSITIVE LOGITS
    vice
    0.31
    äºĭåıij
    0.28
    ector
    0.26
     Resp
    0.25
    寥
    0.25
    pause
    0.24
    quired
    0.24
    ä»ĸå°±
    0.24
    ILITY
    0.23
     MUST
    0.23
    Activations Density 0.589%

    No Known Activations