© 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. 107645
    Prev
    Next
    INDEX
    Explanations

    help/assistance requests

    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
    欲
    -0.33
    éľĢ
    -0.31
    éľĢè¦ģ
    -0.30
    æĪij们è¦ģ
    -0.29
    Ĥ¬
    -0.29
    æĭŁ
    -0.29
    è¿ĺéľĢè¦ģ
    -0.29
    第ä¸ī次
    -0.28
     need
    -0.28
    æĥ³è¦ģ
    -0.28
    POSITIVE LOGITS
    aris
    0.26
    éĮĦ
    0.26
     elo
    0.25
    ascript
    0.25
    è°Ķ
    0.24
    æ¶ĺ
    0.24
    æ¹ĸåĮº
    0.24
     Fundamental
    0.24
     McConnell
    0.23
    æİĩ
    0.23
    Activations Density 1.142%

    No Known Activations