© 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. 30116
    Prev
    Next
    INDEX
    Explanations

    testing and results

    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
    _FINISH
    -0.10
    omi
    -0.10
    çĨ¬
    -0.09
    _FORE
    -0.09
    çĦĻ
    -0.09
     недо
    -0.09
     Doyle
    -0.09
    è¿«
    -0.09
    squ
    -0.08
    éĿł
    -0.08
    POSITIVE LOGITS
    æ£Ģæµĭ
    0.12
    æ£Ģ
    0.12
    檢
    0.11
    plr
    0.10
    gua
    0.10
    æ¤
    0.09
    æ£ĢéªĮ
    0.09
    æĻ®æŁ¥
    0.09
    æĬ½æ£Ģ
    0.09
    çŃĽæŁ¥
    0.09
    Activations Density 0.065%

    No Known Activations