© 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. 7784
    Prev
    Next
    INDEX
    Explanations

    need / want

    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
     Erg
    -0.09
    ilos
    -0.08
     Erot
    -0.08
    çĶ±æѤ
    -0.08
     Horde
    -0.08
     Mane
    -0.08
    综
    -0.08
     Hatch
    -0.08
    engl
    -0.08
    sap
    -0.08
    POSITIVE LOGITS
    ä¹ĭéĻħ
    0.13
    å¯Ħæīĺ
    0.11
    ä¸Ģæľ¬
    0.11
    ä¸Ģ声
    0.10
    åı¯éĿłçļĦ
    0.09
    åį´åıijçİ°
    0.09
    ä¸Ģ棵
    0.09
    å¸Ĥåľºä»½é¢Ŀ
    0.09
    è¨Ĺ
    0.09
    è°ķ
    0.09
    Activations Density 0.029%

    No Known Activations