© 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. 51566
    Prev
    Next
    INDEX
    Explanations

    writes to praise

    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
    çĹ´
    -0.09
    zar
    -0.09
    éºĿ
    -0.09
    ãĥģãĥ³
    -0.09
    ç«
    -0.09
    objectId
    -0.08
    æĪIJå°±
    -0.08
    range
    -0.08
     Arabia
    -0.08
    phans
    -0.08
    POSITIVE LOGITS
    åıįé¦Ī
    0.14
    å°ĸ
    0.11
    ç»ĻäºĨæĪij
    0.10
     feedback
    0.10
    éĢģæĿ¥
    0.10
    ä¸ĬæĿ¥
    0.09
    0.09
    RICT
    0.09
    è¿ĩæĿ¥
    0.09
    ç½ijåıĭ
    0.09
    Activations Density 0.072%

    No Known Activations