© 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
    2. Qwen3-32B
    3. 32-RESID-BATCHTOPK-65K
    4. 51981
    Prev
    Next
    INDEX
    Explanations

    directing actions or flow

    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
    ÑģÑĤеÑĢ
    -0.10
    cz
    -0.09
    çĴĺ
    -0.08
    ocking
    -0.08
    attend
    -0.08
    esco
    -0.08
    atical
    -0.08
    åį¿
    -0.08
    iska
    -0.08
    樵
    -0.08
    POSITIVE LOGITS
    éªł
    0.11
    edly
    0.10
    (Direction
    0.10
    ness
    0.09
    èµ°ä¸ĭåİ»
    0.09
    æĢ§å¼º
    0.09
    人çĶŁçļĦ
    0.09
    èĹģ
    0.09
    etur
    0.08
     Craw
    0.08
    Activations Density 0.039%

    No Known Activations