© 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. 30-TRANSCODER-HP
    7. 96416
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    undle
    -0.26
    让她
    -0.25
    让ä»ĸ
    -0.25
    æĪijåİ»
    -0.25
     cheeks
    -0.25
    _we
    -0.25
    è®ĵä»ĸ
    -0.24
    使ä¹ĭ
    -0.24
    ç§ģãģ¯
    -0.23
    pto
    -0.23
    POSITIVE LOGITS
    ä½łçļĦ
    0.90
    èĩªå·±çļĦ
    0.89
     your
    0.87
    æĤ¨çļĦ
    0.81
    æĪij们çļĦ
    0.78
    æĪijçļĦ
    0.76
    your
    0.73
    äºĨèĩªå·±çļĦ
    0.73
     our
    0.70
     Your
    0.69
    Activations Density 0.901%

    No Known Activations