© 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. 96359
    Prev
    Next
    INDEX
    Explanations

    legal and policy matters

    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
     darkest
    -0.26
    bö
    -0.25
    kest
    -0.25
    åķĨåѦéĻ¢
    -0.24
     FieldType
    -0.24
    *sp
    -0.24
    çĶļèĩ³
    -0.24
     darker
    -0.24
    tré
    -0.24
    borg
    -0.24
    POSITIVE LOGITS
     subject
    0.27
     themselves
    0.26
    é¢ĦåĪ¶
    0.26
    æī©
    0.25
     mer
    0.25
    äºĨåIJ§
    0.25
    æĮ¥æīĭ
    0.25
     tant
    0.25
    æĭ¿æĿ¥
    0.24
    æī©å¤§
    0.24
    Activations Density 0.107%

    No Known Activations