© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52686
    Prev
    Next
    INDEX
    Explanations

    . , rather

    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
    以便
    -0.09
    Ùĩب
    -0.09
    æĦı義
    -0.09
    èĽ®
    -0.08
    @mail
    -0.08
    -days
    -0.08
    ->__
    -0.08
    éĻ¤äºĨ
    -0.08
    azes
    -0.08
    etry
    -0.08
    POSITIVE LOGITS
    çŃīåİŁåĽł
    0.12
    çļĦç¼ĺæķħ
    0.12
    æīĢèĩ´
    0.11
    没åĬŀæ³ķ
    0.10
     alone
    0.10
    åijĹ
    0.10
     reasons
    0.10
    主è¦ģåİŁåĽł
    0.10
     reason
    0.10
    没æ³ķ
    0.10
    Activations Density 0.201%

    No Known Activations