© 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. 108398
    Prev
    Next
    INDEX
    Explanations

    of

    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
    ä¹ĭæīĢ
    -0.30
    azu
    -0.29
    lad
    -0.28
    uin
    -0.27
    неÑģен
    -0.27
    æŃ£æĺ¯
    -0.26
    çļĦ羣å®ŀ
    -0.26
    bart
    -0.25
    avy
    -0.25
    ko
    -0.25
    POSITIVE LOGITS
    éĹ
    0.28
    ...</
    0.26
    odont
    0.26
    éĹ³
    0.25
    ()</
    0.25
    "<?
    0.25
    _SCHEDULE
    0.25
     anywhere
    0.24
    éĨļ
    0.24
    ç»į
    0.24
    Activations Density 1.095%

    No Known Activations