© 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. 23562
    Prev
    Next
    INDEX
    Explanations

    otherwise + conjunction/adverb

    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
    itud
    -0.10
    itious
    -0.09
    shot
    -0.09
    讶
    -0.09
    è¾ķ
    -0.09
    鹧
    -0.09
     zwar
    -0.09
    shots
    -0.09
    ificant
    -0.09
    STER
    -0.09
    POSITIVE LOGITS
    elves
    0.11
    eper
    0.10
    ull
    0.09
    æĥħåĨµä¸ĭ
    0.09
    chestra
    0.09
    еÑģÑĤво
    0.09
    ector
    0.09
    levance
    0.09
    ep
    0.08
     Lamar
    0.08
    Activations Density 0.010%

    No Known Activations