© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 15305
    Prev
    Next
    INDEX
    Explanations

    danger and hazards

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĻĥ
    -0.20
    å¨ĵ
    -0.19
    ussy
    -0.19
    ç¢ĩ
    -0.19
    çĦķ
    -0.19
    åķ¶
    -0.19
    线æĿ¡
    -0.18
     undermin
    -0.18
     cumshot
    -0.18
     disappe
    -0.18
    POSITIVE LOGITS
    éļľç¢į
    0.29
    éĻ·éĺ±
    0.29
    åį±éĻ©
    0.27
     dangers
    0.26
    éļľ
    0.25
     obstacles
    0.24
    lest
    0.23
    伺
    0.23
    éĺ»ç¢į
    0.23
     danger
    0.23
    Activations Density 0.053%

    No Known Activations