© 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. 51881
    Prev
    Next
    INDEX
    Explanations

    actual size or number

    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
    amt
    -0.10
    便
    -0.08
    .setTimeout
    -0.08
    çĥ¦
    -0.07
    .dom
    -0.07
    orea
    -0.07
    ellung
    -0.07
    Mocks
    -0.07
    ammu
    -0.07
    Caught
    -0.07
    POSITIVE LOGITS
    å®ŀéĻħ
    0.20
    çļĦå®ŀéĻħ
    0.18
    實éļĽ
    0.15
     actual
    0.15
     ìĭ¤ìłľ
    0.14
    å®ŀéĻħæĥħåĨµ
    0.14
    actual
    0.14
    	actual
    0.13
    Actual
    0.13
    (actual
    0.13
    Activations Density 0.103%

    No Known Activations