© 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. 20007
    Prev
    Next
    INDEX
    Explanations

    pen or pencil related words

    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
    æĹ¥åŃIJ
    -0.10
    éĥ¢
    -0.10
    å¼±
    -0.10
    odia
    -0.09
    æīģ
    -0.09
    weak
    -0.09
     Albany
    -0.09
    avar
    -0.09
    åij±
    -0.09
     weak
    -0.09
    POSITIVE LOGITS
     pen
    0.14
    ç¬Ķ
    0.12
    pen
    0.12
     Pen
    0.10
     pens
    0.10
    shit
    0.10
    éĢłçº¸
    0.10
    èĻ±
    0.10
    Pen
    0.09
    å¢ħ
    0.09
    Activations Density 0.011%

    No Known Activations