© 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. 20877
    Prev
    Next
    INDEX
    Explanations

    steroids

    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
    åĩ«
    -0.09
    å¥ļ
    -0.09
    quil
    -0.08
    ixo
    -0.08
    çĦ¦
    -0.08
     pix
    -0.08
    ixin
    -0.08
     khu
    -0.08
     Feinstein
    -0.08
     unc
    -0.08
    POSITIVE LOGITS
    ACHE
    0.10
     uv
    0.09
    é¬
    0.09
    ä¸įçĶĺ
    0.09
    å»ĵ
    0.09
    å¼ĤæĢ§
    0.09
    è½®
    0.09
    èµĦæľ¬
    0.08
    ç´«å¤ĸ线
    0.08
    è¾ĥéĩı
    0.08
    Activations Density 0.379%

    No Known Activations