© 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. 21443
    Prev
    Next
    INDEX
    Explanations

    original article

    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
    èĩ§
    -0.09
    Mos
    -0.09
    çĸ¤
    -0.09
    ä¸ĵéŨ
    -0.08
    ä¹ĭåĨħ
    -0.08
    oter
    -0.08
    çĦī
    -0.08
     fulfillment
    -0.08
    è°·
    -0.07
    è¯Ħå§Ķ
    -0.07
    POSITIVE LOGITS
    åİŁæĸĩ
    0.15
     original
    0.13
    /original
    0.13
    (original
    0.12
    original
    0.12
     оÑĢигина
    0.11
    éĿ¢çĽ®
    0.11
     Original
    0.11
    åİŁèijĹ
    0.11
     originals
    0.11
    Activations Density 0.064%

    No Known Activations