© 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. 26530
    Prev
    Next
    INDEX
    Explanations

    introduction section start

    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
    ãģĹãģ¾ãģĦ
    -0.12
    /or
    -0.11
    ãģĹãģ¾ãģĨ
    -0.11
    ëģĶ
    -0.11
    ieren
    -0.10
    بÙĥ
    -0.10
    çįŃ
    -0.10
    ãģĭãģ¤
    -0.09
    èĮ§
    -0.09
    oval
    -0.09
    POSITIVE LOGITS
    ../
    0.14
    ties
    0.14
    èĩ¾
    0.12
    istic
    0.11
    ision
    0.11
    ents
    0.11
    ãĤĪãģĨãģª
    0.11
    nection
    0.10
    ial
    0.10
    ially
    0.10
    Activations Density 0.069%

    No Known Activations