© 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. 30841
    Prev
    Next
    INDEX
    Explanations

    website, official, .com

    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
    akis
    -0.09
     fade
    -0.08
    lation
    -0.08
     commun
    -0.08
     mus
    -0.08
    locking
    -0.07
    _Form
    -0.07
     ÐĴлад
    -0.07
    .typ
    -0.07
    hal
    -0.07
    POSITIVE LOGITS
    æŃĩ
    0.11
    åĥŃ
    0.10
    åıijæĺİ
    0.09
    å°Ĩé¢Ĩ
    0.09
     Tong
    0.09
    å¸ĤåĢ¼
    0.08
    çĻ½é¢Ĩ
    0.08
    æĸĩåĮĸ建设
    0.08
     hit
    0.08
    åĪĽæĸ°èĥ½åĬĽ
    0.08
    Activations Density 0.033%

    No Known Activations