© 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. 31-TRANSCODER-HP
    7. 18474
    Prev
    Next
    INDEX
    Explanations

    1 or 2

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    emu
    -0.29
    омÑĥ
    -0.28
     ÐľÐ¾ÑģковÑģк
    -0.28
    è¯ĬæĸŃ
    -0.27
    Ñĥ
    -0.26
    imu
    -0.26
    isu
    -0.26
    __,__
    -0.25
    å¹³æ°ij
    -0.25
    arness
    -0.25
    POSITIVE LOGITS
    大ä¼ļä¸Ĭ
    0.33
    æľŁåĨħ
    0.29
    人群ä¸Ń
    0.27
    è¡Įä¸ļä¸Ń
    0.27
    èĮĥåĽ´åĨħ
    0.27
     shy
    0.25
    çĹĤ
    0.25
    èĬĤ缮ä¸Ń
    0.24
    SAFE
    0.24
    biased
    0.24
    Activations Density 0.051%

    No Known Activations