© 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. 52819
    Prev
    Next
    INDEX
    Explanations

    water molecules

    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
    cean
    -0.10
    ails
    -0.09
    æķ´ä¸ªäºº
    -0.09
    zac
    -0.09
    iant
    -0.09
    å±ķ
    -0.09
    æ¡Ī
    -0.08
    Expert
    -0.08
    æĮ¯
    -0.08
    æĢ»ä½ĵè§ĦåĪĴ
    -0.08
    POSITIVE LOGITS
    åıĬåħ¶ä»ĸ
    0.10
    æ¹ľ
    0.09
    å°ıæŶ
    0.09
    -gnu
    0.09
    åłĥ
    0.08
    åĨ°
    0.08
    ðŁĴ§
    0.08
    æĺ¯æľĢ
    0.08
    usal
    0.08
    .camel
    0.08
    Activations Density 0.190%

    No Known Activations