© 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. 24-TRANSCODER-HP
    7. 150910
    Prev
    Next
    INDEX
    Explanations

    twenty

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    енÑĤа
    -0.27
    orget
    -0.27
    rita
    -0.25
    reno
    -0.25
    /*@
    -0.25
    幺
    -0.25
    éŃħ
    -0.24
     charm
    -0.24
    æģ¬
    -0.24
    orent
    -0.24
    POSITIVE LOGITS
    éģĵ
    0.29
    th
    0.29
    0
    0.28
    outs
    0.28
    æľĢæĹ©çļĦ
    0.28
    éĤ£éĩĮ
    0.27
    éĤ£ä¸ªäºº
    0.27
    AF
    0.27
    B
    0.26
    af
    0.26
    Activations Density 0.866%

    No Known Activations