© 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. 108224
    Prev
    Next
    INDEX
    Explanations

    it

    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
    _marshall
    -0.27
    客æĪ·æľįåĬ¡
    -0.27
     constit
    -0.26
    subtype
    -0.26
    èĢ³
    -0.25
    è¾Ł
    -0.25
     repealed
    -0.25
     capit
    -0.24
    åŃ©
    -0.24
    favor
    -0.24
    POSITIVE LOGITS
    ä¸ĩ亿
    0.27
    åĪ¶éĢłä¸ļ
    0.24
    _increment
    0.24
     Inline
    0.24
     tun
    0.23
     currently
    0.23
     radicals
    0.23
    é«ĺ温
    0.23
     really
    0.23
    prech
    0.23
    Activations Density 0.934%

    No Known Activations