© 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. 18909
    Prev
    Next
    INDEX
    Explanations

    relative position or quantity

    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
    两个
    -0.31
    dojo
    -0.27
    olare
    -0.27
    åħ©åĢĭ
    -0.27
    Æ¡i
    -0.25
    _WH
    -0.25
    /big
    -0.25
    ä½İ
    -0.24
    ANCH
    -0.24
    对èĩªå·±çļĦ
    -0.24
    POSITIVE LOGITS
     approximately
    0.39
    0.36
     fifty
    0.34
    ä¸ĢçĻ¾å¤ļ
    0.33
    大约
    0.33
    ä¸ĢçĻ¾
    0.32
    äºĶåįģ
    0.32
     +
    0.31
    äºĶçĻ¾
    0.30
    +
    0.29
    Activations Density 1.472%

    No Known Activations