© 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. 96281
    Prev
    Next
    INDEX
    Explanations

    Non-English names or locations

    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
    éĹ®çŃĶ
    -0.35
    å¡«æĬ¥
    -0.26
    éĶ»
    -0.26
    éĹ®è¯¢
    -0.25
     Threat
    -0.25
    æĥħ
    -0.25
    con
    -0.24
    RuleContext
    -0.24
    threat
    -0.24
    è¿ŀèĥľ
    -0.24
    POSITIVE LOGITS
    æij¸
    0.28
     поÑģл
    0.26
     retal
    0.25
    zer
    0.25
    å°ı康社ä¼ļ
    0.25
     pard
    0.24
    credible
    0.24
    æł¹æľ¬
    0.24
    æ§ģ
    0.24
    IELDS
    0.24
    Activations Density 0.103%

    No Known Activations