© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 30979
    Prev
    Next
    INDEX
    Explanations

    interesting thing/tidbit/numbers/assignments/cards/round/premise

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    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
    åŃIJéĩĮ
    -0.10
    æĥħæĻ¯
    -0.10
    çľ¼æ³ª
    -0.10
    ť
    -0.10
    å¯Įè´µ
    -0.09
    akah
    -0.09
    xl
    -0.08
    æĺŁè¾°
    -0.08
    æľ
    -0.08
    éĿ¡
    -0.08
    POSITIVE LOGITS
    ly
    0.16
    LY
    0.11
    ness
    0.11
    ment
    0.10
    è¨ĢãģĪãģ°
    0.10
    -looking
    0.10
    çĽİ
    0.09
    çĪ±å¥½
    0.09
    entials
    0.09
    thal
    0.09
    Activations Density 0.011%

    No Known Activations