© 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. 22009
    Prev
    Next
    INDEX
    Explanations

    Halloween and related suffixes

    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
    erval
    -0.10
    zÄħ
    -0.10
    YPE
    -0.10
    çĸ²åĬ³
    -0.09
    ç¾İæľ¯é¦Ĩ
    -0.09
    ç±³åħ°
    -0.09
    IJèĹı
    -0.09
    иÑĢовка
    -0.09
    们çļĦ
    -0.09
    lob
    -0.09
    POSITIVE LOGITS
    itas
    0.09
    ish
    0.09
    ism
    0.09
    aly
    0.08
    bero
    0.08
    -themed
    0.08
    opus
    0.08
    atal
    0.08
    jin
    0.08
    ellan
    0.08
    Activations Density 0.006%

    No Known Activations