© 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. 20-TRANSCODER-HP
    7. 129730
    Prev
    Next
    INDEX
    Explanations

    political discussions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    phoon
    -0.27
    vation
    -0.25
     penc
    -0.25
    æ±²
    -0.25
    绣
    -0.25
    æķ
    -0.25
    (rect
    -0.24
     Eyl
    -0.24
    åĬłæ²¹
    -0.24
    ä¸Ģäºĭ
    -0.24
    POSITIVE LOGITS
    空åīį
    0.28
     veget
    0.27
    GES
    0.27
     turnovers
    0.26
    Turning
    0.25
    acky
    0.24
    å¢ĥå¤ĸ
    0.24
     Congressional
    0.24
     realistically
    0.24
     stray
    0.24
    Activations Density 17.086%

    No Known Activations