© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8114
    Prev
    Next
    INDEX
    Explanations

    Webster /library

    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
    èĬĴ
    -0.09
    orem
    -0.09
    çĦ¶æĺ¯
    -0.08
    acha
    -0.08
    Ñĩив
    -0.08
    aines
    -0.08
    ingly
    -0.08
    amental
    -0.08
    aces
    -0.08
    ima
    -0.08
    POSITIVE LOGITS
    à¹Ħà¸ĭ
    0.11
    lio
    0.09
    ights
    0.09
    é»ijå¤ľ
    0.09
    roids
    0.09
    ORTH
    0.09
    ities
    0.09
    ijn
    0.09
    ilder
    0.09
    æĻŀ
    0.09
    Activations Density 0.002%

    No Known Activations