© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51269
    Prev
    Next
    INDEX
    Explanations

    outcomes and results

    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
     cro
    -0.10
    èĸĦå¼±
    -0.09
    .ib
    -0.09
    æİĴ
    -0.09
    yne
    -0.08
    cro
    -0.08
    绣é¢Ĩ
    -0.08
    née
    -0.08
    iesta
    -0.08
    .Argument
    -0.08
    POSITIVE LOGITS
    ç»ĵå±Ģ
    0.36
     outcome
    0.31
     outcomes
    0.28
    outcome
    0.28
    Outcome
    0.28
     Outcome
    0.26
    ç»ĵæŀľ
    0.19
    çļĦç»ĵæŀľ
    0.19
    çµIJæŀľ
    0.17
     ê²°ê³¼
    0.16
    Activations Density 0.222%

    No Known Activations