© 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. 27750
    Prev
    Next
    INDEX
    Explanations

    porous and poro- prefixes

    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
    -0.10
    ãģĭãĤĭ
    -0.09
    /pi
    -0.09
     addCriterion
    -0.09
    ourney
    -0.09
    полн
    -0.09
    ваÑĢ
    -0.09
    ISTR
    -0.09
    ãĤıãĤĬ
    -0.09
    ufig
    -0.09
    POSITIVE LOGITS
    centaje
    0.15
    celain
    0.13
    centage
    0.12
     lẽ
    0.12
    èĩ¾
    0.11
    iento
    0.11
    onto
    0.10
    ignal
    0.10
    cel
    0.10
    folio
    0.10
    Activations Density 0.036%

    No Known Activations