© 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. 108556
    Prev
    Next
    INDEX
    Explanations

    geological terrain

    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
    ë¦Ħ
    -0.26
    åı¯æĥ³
    -0.25
    ingles
    -0.25
    peech
    -0.24
    hort
    -0.24
    anse
    -0.24
    iode
    -0.24
    èĴī
    -0.24
     rightly
    -0.24
    责任åĪ¶
    -0.23
    POSITIVE LOGITS
    æ·ĭ
    0.28
    -ring
    0.26
     ~(
    0.26
    èįĨå·ŀ
    0.26
    åľ°æ®µ
    0.25
    对éĿ¢
    0.25
    ventory
    0.24
    Inventory
    0.24
     pudd
    0.24
    geb
    0.24
    Activations Density 0.919%

    No Known Activations