© 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. 18619
    Prev
    Next
    INDEX
    Explanations

    there

    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
     last
    -0.43
    (last
    -0.38
    last
    -0.36
    -last
    -0.35
    [last
    -0.35
    ,last
    -0.34
    	last
    -0.33
    .last
    -0.32
    æľĢå¾Į
    -0.31
    >Last
    -0.31
    POSITIVE LOGITS
     Having
    0.44
    Having
    0.42
     having
    0.42
    having
    0.39
    å¤ļä¹Ī
    0.28
    æľīäºĨ
    0.28
    boxes
    0.28
    ä¸įå¼Ģ
    0.28
    ç»ĨåĪĨ
    0.27
    æľīçĿĢ
    0.27
    Activations Density 1.143%

    No Known Activations