© 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. 9264
    Prev
    Next
    INDEX
    Explanations

    electronics and framework components

    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
    ote
    -0.08
    òng
    -0.08
    .define
    -0.08
    åıijçİ°éĹ®é¢ĺ
    -0.08
     Natal
    -0.08
     clo
    -0.08
    oven
    -0.07
    anos
    -0.07
    иÑĤел
    -0.07
     pacman
    -0.07
    POSITIVE LOGITS
    aten
    0.10
    _locals
    0.10
    æĪijåĸľæ¬¢
    0.10
     subsequ
    0.09
    éºŁ
    0.09
    å·¡åĽŀ
    0.09
    人群ä¸Ń
    0.09
    çīĽå¸Ĥ
    0.09
    acent
    0.09
    éķ¯
    0.09
    Activations Density 0.005%

    No Known Activations