© 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. 40846
    Prev
    Next
    INDEX
    Explanations

    select

    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
    广
    -0.30
     broad
    -0.27
    olith
    -0.26
    çIJĨæĥ³ä¿¡å¿µ
    -0.25
    廣
    -0.24
     Broad
    -0.24
    çois
    -0.24
    trer
    -0.24
     Lim
    -0.24
    axed
    -0.24
    POSITIVE LOGITS
     sn
    0.27
     dragon
    0.26
     swaps
    0.26
    _OD
    0.26
    notated
    0.25
    worksheet
    0.25
    注æĺİåĩºå¤Ħ
    0.24
     dragons
    0.24
    秾
    0.24
    è¿Ľ
    0.23
    Activations Density 0.328%

    No Known Activations