© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 3095
    Prev
    Next
    INDEX
    Explanations

    handed to, out, over, down, off

    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
    inine
    -0.09
    vil
    -0.09
    hores
    -0.08
    åħ³äºİæĪij们
    -0.08
    ÙĦÙĬÙĥ
    -0.08
    нов
    -0.08
    说æĺİ
    -0.08
    ooks
    -0.08
    inning
    -0.08
    ansk
    -0.08
    POSITIVE LOGITS
    åĽŀæĶ¶
    0.09
    ration
    0.09
    éĴī
    0.09
    ãģ¨åħ±
    0.08
    好å¤Ħ
    0.08
    Õ¦
    0.08
    AMD
    0.08
    iani
    0.08
    _singleton
    0.08
    arking
    0.08
    Activations Density 0.005%

    No Known Activations