© 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. 9884
    Prev
    Next
    INDEX
    Explanations

    identification numbers and tracing

    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
    _Common
    -0.09
    æ®ĵ
    -0.09
    åĽ«
    -0.09
    ?action
    -0.09
    é¦Ĩ
    -0.08
    éĢĨè¡Į
    -0.08
    ä¿®
    -0.08
    ä¹Ĵä¹ĵ
    -0.08
     adjud
    -0.08
    éĩijå±±
    -0.08
    POSITIVE LOGITS
    çĥĻ
    0.12
    æłĩè®°
    0.12
    æłĩè¯Ĩ
    0.11
    åį°è®°
    0.11
    éĵŃ
    0.10
    èĢ»
    0.09
    visible
    0.09
     stamped
    0.09
    å½Ĵå±ŀ
    0.09
    马åħĭ
    0.09
    Activations Density 0.093%

    No Known Activations