© 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. 108399
    Prev
    Next
    INDEX
    Explanations

    )

    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
    Tit
    -0.28
     Tit
    -0.28
    ãĥ¼ãĤº
    -0.26
     zostaÄĩ
    -0.26
    duÄŁ
    -0.26
     tit
    -0.24
     fri
    -0.24
    æĪ¢
    -0.24
    tit
    -0.24
    ëĭĿ
    -0.24
    POSITIVE LOGITS
    她æĺ¯
    0.28
    umann
    0.26
     example
    0.26
    otta
    0.26
    exemple
    0.25
    é¢ī
    0.25
    åĨįçĶŁ
    0.25
    å®ĥæĺ¯
    0.24
    èĥĮä¸Ĭ
    0.24
    itational
    0.24
    Activations Density 0.968%

    No Known Activations