© 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. 18540
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    æīĽ
    -0.25
    ãĥıãĤ¦
    -0.25
    _pwm
    -0.24
    èľ¡
    -0.24
    åIJİç»Ń
    -0.24
    踩
    -0.24
    ?v
    -0.24
    姿åĬ¿
    -0.24
    (Member
    -0.24
    fault
    -0.23
    POSITIVE LOGITS
     полов
    0.27
     slee
    0.26
    -bel
    0.26
    amax
    0.26
    åijĬè¯ī她
    0.25
    Han
    0.24
    DDD
    0.24
    ych
    0.24
     Hil
    0.23
    her
    0.23
    Activations Density 0.002%

    No Known Activations

    This feature has no known activations.