© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 14023
    Prev
    Next
    INDEX
    Explanations

    hexahydro and tetrahydro

    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
     stall
    -0.23
     rag
    -0.21
     stick
    -0.21
    è¶ħ
    -0.19
     pref
    -0.19
    çŀŃ
    -0.19
     PREF
    -0.19
     sil
    -0.18
    æĻ¾
    -0.18
     tz
    -0.18
    POSITIVE LOGITS
    è¯Ĥ
    0.22
    ä¸įæĺİçĻ½
    0.20
    ret
    0.20
    éħ®
    0.20
    饱满
    0.19
    æ¦ĸ
    0.19
    æ²³æµģåŁŁ
    0.19
    活泼
    0.19
    locking
    0.19
    礼æĭľ
    0.18
    Activations Density 0.123%

    No Known Activations