© 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. 18188
    Prev
    Next
    INDEX
    Explanations

    tension

    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
    æī£
    -0.35
    éĵ¾æĿ¡
    -0.35
     link
    -0.34
    éĵ¾æİ¥
    -0.34
    link
    -0.33
    æİ¥
    -0.33
     links
    -0.32
    çݯ
    -0.31
     chains
    -0.31
    éĹŃ
    -0.31
    POSITIVE LOGITS
     tensions
    0.42
     tension
    0.42
     tens
    0.34
    ãĥĨãĥ³
    0.33
     tight
    0.32
    -t
    0.32
    ç´§å¼ł
    0.32
    æ¯Ķäºļ
    0.29
     Ten
    0.29
     ten
    0.29
    Activations Density 11.646%

    No Known Activations