© 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. 107782
    Prev
    Next
    INDEX
    Explanations

    bl

    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
    ICLE
    -0.33
    ials
    -0.30
    -domain
    -0.29
    icle
    -0.28
     those
    -0.28
    ionales
    -0.27
    iox
    -0.26
    _exports
    -0.26
    iards
    -0.24
    iren
    -0.24
    POSITIVE LOGITS
    å¹³åı°ä¸Ĭ
    0.26
    uder
    0.26
     Haram
    0.24
    ahir
    0.24
    èĩªçĦ¶èĢĮ
    0.24
    平淡
    0.24
    è¿Ļåıªæĺ¯
    0.24
    è£ģåĪ¤
    0.23
     Curtain
    0.23
     horrors
    0.23
    Activations Density 1.566%

    No Known Activations