© 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. 14-TRANSCODER-HP
    7. 41026
    Prev
    Next
    INDEX
    Explanations

    help

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_14.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.14.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    个ä½ĵ
    -0.27
    å±ħ室
    -0.27
    ardin
    -0.26
    éĩĩ访æŶ
    -0.26
    çĸµ
    -0.24
    cell
    -0.24
    attend
    -0.24
     thá»ķ
    -0.24
    tparam
    -0.24
    æŁIJ个
    -0.24
    POSITIVE LOGITS
     Shak
    0.28
    ä¹ĭ乡
    0.27
    èĩ³ä¸Ĭ
    0.26
     pr
    0.26
     suce
    0.25
    Wh
    0.25
    KEN
    0.24
    /Sh
    0.24
    allah
    0.24
     Wh
    0.24
    Activations Density 1.664%

    No Known Activations