© 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. 108193
    Prev
    Next
    INDEX
    Explanations

    Names and foreign languages

    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
    ipes
    -0.26
    atsu
    -0.25
    ÃŃt
    -0.25
    å°¼åħĭ
    -0.25
    UMP
    -0.24
    iso
    -0.24
    ÃŃv
    -0.24
    ilies
    -0.24
    oti
    -0.24
    alan
    -0.24
    POSITIVE LOGITS
    çĶŁäº§åĴĮ
    0.27
    è¡įçĶŁ
    0.27
    è¡į
    0.26
    åĮ¿åIJį
    0.26
    çģ¯ç¬¼
    0.25
    ç¼Ī
    0.25
    ctl
    0.25
    延伸
    0.25
    endregion
    0.25
    ลà¸ĩ
    0.24
    Activations Density 2.145%

    No Known Activations