© 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. 107844
    Prev
    Next
    INDEX
    Explanations

    updated or added

    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
    PERT
    -0.30
    ç«Ļåľ¨
    -0.28
    欢è¿İæĤ¨
    -0.26
    æĻºåºĵ
    -0.26
     Homepage
    -0.26
    å¿«æĬ¥
    -0.25
    âŀ¤
    -0.25
    erable
    -0.25
    ily
    -0.25
    çļĦç¡®æĺ¯
    -0.25
    POSITIVE LOGITS
    bi
    0.28
    _secs
    0.27
     resembling
    0.26
     bi
    0.25
     munic
    0.25
    åıĬæŶ
    0.24
    ours
    0.24
     relief
    0.24
     discussed
    0.23
    éĻĨç»Ń
    0.23
    Activations Density 0.910%

    No Known Activations