© 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. 37508
    Prev
    Next
    INDEX
    Explanations

    more / most / best / increasingly

    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
    产
    -0.10
    æľīéĴ±
    -0.08
    æĺ§
    -0.08
    etta
    -0.08
    lew
    -0.08
    åİī害
    -0.08
    oon
    -0.08
    çĶļ
    -0.08
    éĢŁåº¦å¿«
    -0.08
    æĹº
    -0.08
    POSITIVE LOGITS
     more
    0.14
    æĽ´éĩįè¦ģ
    0.13
    æĽ´
    0.13
    æĽ´åħ·
    0.12
    æĽ´ä¸º
    0.12
    æĽ´æľī
    0.12
    æĽ´éĩįè¦ģçļĦ
    0.11
     более
    0.11
    _more
    0.11
    indle
    0.11
    Activations Density 0.087%

    No Known Activations