© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20576
    Prev
    Next
    INDEX
    Explanations

    dangerous

    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
    çĶ¬
    -0.13
    cie
    -0.10
    zilla
    -0.10
    åŃIJéĩĮ
    -0.09
    stone
    -0.09
    ../../../
    -0.09
    istic
    -0.09
    stones
    -0.09
    antan
    -0.08
    checker
    -0.08
    POSITIVE LOGITS
    ness
    0.17
    -looking
    0.15
    NESS
    0.12
    ly
    0.09
    ecz
    0.09
    éĢĨ
    0.09
    uly
    0.09
    å¾Ĵ
    0.09
    åľ°å¸¦
    0.09
    LY
    0.09
    Activations Density 0.006%

    No Known Activations