© 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. 35584
    Prev
    Next
    INDEX
    Explanations

    counting letters or 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
    éĵº
    -0.09
    è¿Ļ两ç§į
    -0.09
    äºĮèĢħ
    -0.09
    çİ
    -0.09
    æŃ£è§Ħ
    -0.08
     пÑĢеж
    -0.08
    延伸
    -0.08
     Foley
    -0.08
    extend
    -0.08
    èĬ¥
    -0.08
    POSITIVE LOGITS
     letters
    0.12
    letter
    0.11
    åįķè¯į
    0.11
    letters
    0.10
     letter
    0.10
    åŃĹæ¯į
    0.10
    åĨĻåŃĹ
    0.10
    éĢĨåĬ¿
    0.10
    uchar
    0.09
    Letter
    0.09
    Activations Density 0.167%

    No Known Activations