© 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. 150283
    Prev
    Next
    INDEX
    Explanations

    question and problem

    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
    æīĵå¼Ģ
    -0.27
    åĥı个
    -0.26
    rowsing
    -0.26
    æ¸IJ
    -0.26
    “(
    -0.25
     Scenes
    -0.25
    emos
    -0.24
    å°±å¤ŁäºĨ
    -0.24
    ãģĤãģ£ãģŁ
    -0.23
    å°±æľīäºĨ
    -0.23
    POSITIVE LOGITS
     aims
    0.29
     assumes
    0.29
     specifies
    0.28
     defines
    0.27
    ãĤ¼
    0.26
    _descriptor
    0.25
     deals
    0.25
     manip
    0.25
     splits
    0.25
     statement
    0.25
    Activations Density 0.464%

    No Known Activations