© 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. 49127
    Prev
    Next
    INDEX
    Explanations

    corresponding to

    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
    æ»ĵ
    -0.13
    -0.09
    igh
    -0.09
    idge
    -0.09
    ¶
    -0.09
    èĻ±
    -0.09
    izia
    -0.09
    amente
    -0.08
    åij¦
    -0.08
    é«ĺåİŁ
    -0.08
    POSITIVE LOGITS
    ents
    0.10
    -Mart
    0.10
    ingly
    0.10
    uddle
    0.10
    ences
    0.09
    uld
    0.09
    minimal
    0.09
    å±Ĭæ¯ķä¸ļçĶŁ
    0.08
    olean
    0.08
    ermal
    0.08
    Activations Density 0.022%

    No Known Activations