© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 62461
    Prev
    Next
    INDEX
    Explanations

    aspect ratio

    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
    ç´¢
    -0.09
    åĬŁå¤«
    -0.09
     pin
    -0.08
    èĥĮåĮħ
    -0.08
    зн
    -0.08
    éģIJ
    -0.08
     Bare
    -0.07
    éľĩ
    -0.07
    æ¡Ī
    -0.07
    èĩªåıij
    -0.07
    POSITIVE LOGITS
    æ¯Ķçİĩ
    0.13
     ratio
    0.12
    æ¯Ķä¾ĭ
    0.12
    ratio
    0.12
    失è°ĥ
    0.11
    æ¯ĶåĪ©
    0.11
    Ratio
    0.10
     Ratio
    0.10
     ratios
    0.10
    _ratio
    0.10
    Activations Density 0.038%

    No Known Activations