© 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. 33-TRANSCODER-HP
    7. 53814
    Prev
    Next
    INDEX
    Explanations

    running for election

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     trif
    -0.31
    è·ij
    -0.31
    ley
    -0.29
     runner
    -0.28
     ele
    -0.28
    è·ijäºĨ
    -0.27
     campaign
    -0.27
     race
    -0.26
     campaña
    -0.26
    é
    -0.26
    POSITIVE LOGITS
    snd
    0.27
    æķ¢äºİ
    0.26
    ubit
    0.25
    bang
    0.25
    ?=.*
    0.25
    oppable
    0.25
    çģŀ
    0.24
    ihad
    0.24
     quyá»ģn
    0.23
    /fw
    0.23
    Activations Density 0.015%

    No Known Activations