© 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. 8637
    Prev
    Next
    INDEX
    Explanations

    rest of

    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
    ãģ¹ãģį
    -0.10
    ãĤº
    -0.10
    ÐŀÐĿ
    -0.10
    STER
    -0.09
    å®ł
    -0.09
    гÑĥ
    -0.09
    ssc
    -0.09
    é¹ħ
    -0.09
    ect
    -0.09
    ãģ¦
    -0.09
    POSITIVE LOGITS
    orative
    0.14
    aurants
    0.12
    lessness
    0.12
    RICT
    0.12
    itution
    0.10
    OfString
    0.10
    udio
    0.09
    ocking
    0.09
    udy
    0.09
    ocked
    0.09
    Activations Density 0.020%

    No Known Activations