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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51688
    Prev
    Next
    INDEX
    Explanations

    point of view

    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
    rule
    -0.09
    è¯Ń
    -0.09
    rien
    -0.09
    车祸
    -0.09
    äºĨä¸Ģæī¹
    -0.08
    Rule
    -0.08
    颧
    -0.08
    æĢ§æł¼
    -0.08
    æĦıä¹īä¸Ĭ
    -0.08
    蹶
    -0.08
    POSITIVE LOGITS
    éļĶçĿĢ
    0.10
    FLT
    0.09
    庶
    0.09
    åģľçķĻ
    0.09
    ignum
    0.09
    ÑĤек
    0.09
    .ordinal
    0.09
    ì°°
    0.08
    æĹģ
    0.08
    fresh
    0.08
    Activations Density 0.070%

    No Known Activations