© 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. 25-TRANSCODER-HP
    7. 108660
    Prev
    Next
    INDEX
    Explanations

    Business reports and disclosures

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĬĺå°Ħ
    -0.29
    ä¸Ģåı¥è¯Ŀ
    -0.27
    为ä¸Ģä½ĵ
    -0.27
    好äºĭ
    -0.27
    好çļĦ
    -0.26
    urus
    -0.25
    é¤IJé¦Ĩ
    -0.24
    å°±éľĢè¦ģ
    -0.24
    çļĦä¸Ģ个
    -0.24
    моÑĤ
    -0.24
    POSITIVE LOGITS
     compensation
    0.28
    educt
    0.25
    esz
    0.25
    ateral
    0.25
    åĹij
    0.25
    Method
    0.24
    erial
    0.24
    _crossentropy
    0.24
     certain
    0.24
     assumptions
    0.24
    Activations Density 0.910%

    No Known Activations