© 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. 55756
    Prev
    Next
    INDEX
    Explanations

    preservation

    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
    arry
    -0.10
    /or
    -0.09
    adesh
    -0.09
    åħĴ
    -0.09
    ån
    -0.09
    å¯Įæľī
    -0.09
    çĢij
    -0.09
    STER
    -0.09
    ipo
    -0.08
    aro
    -0.08
    POSITIVE LOGITS
    umably
    0.10
    ä¸İåıijå±ķ
    0.10
     vững
    0.09
    entially
    0.09
    Ñħ
    0.09
    asive
    0.09
    ierung
    0.09
    ably
    0.09
    ista
    0.09
    isted
    0.09
    Activations Density 0.018%

    No Known Activations