© 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. Qwen3-8B
    5. 18-RESID-BATCHTOPK-65K__L0-80
    7. 2192
    Prev
    Next
    INDEX
    Explanations

    consistent [word]

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-8b-saes/saes_Qwen_Qwen3-8B_batch_top_k/resid_post_layer_18
    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
    æѤåĪ»
    -0.22
    noon
    -0.21
    对çŃĸ
    -0.20
    nar
    -0.20
    abh
    -0.20
    /her
    -0.19
    ISTER
    -0.19
    iced
    -0.18
    onet
    -0.18
    ISR
    -0.18
    POSITIVE LOGITS
    æĢ§åĴĮ
    0.36
    æĢ§çļĦ
    0.36
    æĢ§å¼º
    0.30
    æĢ§
    0.28
    ently
    0.26
    emente
    0.26
     throughout
    0.24
     uptime
    0.24
    -minded
    0.24
    çł¥
    0.23
    Activations Density 0.021%

    No Known Activations