© 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. 23599
    Prev
    Next
    INDEX
    Explanations

    that followed by a clause or phrase

    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
    æ°¸ä¹ħ
    -0.10
    ä½ľä¸ºä¸Ģç§į
    -0.09
    ä½ľä¸ºä¸ĢåIJį
    -0.09
    为代表
    -0.09
    vak
    -0.09
    _NV
    -0.08
    å½ĵçĦ¶
    -0.08
    æĪĸ許
    -0.08
    æĪĸ许æĺ¯
    -0.08
    éļħ
    -0.08
    POSITIVE LOGITS
    è¶Ĭæĺ¯
    0.12
    2
    0.10
    7
    0.10
    6
    0.09
    0
    0.09
    1
    0.09
    åIJİæĤĶ
    0.09
    3
    0.09
    模èĮĥ
    0.09
    çĥ
    0.09
    Activations Density 0.289%

    No Known Activations