© 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. 62040
    Prev
    Next
    INDEX
    Explanations

    ending in ston or ton

    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
    zzo
    -0.09
    ums
    -0.08
    tones
    -0.08
    æķħ乡
    -0.08
    oo
    -0.08
    enal
    -0.08
    thic
    -0.08
    anst
    -0.08
    sert
    -0.08
    åįģ足
    -0.08
    POSITIVE LOGITS
    ablish
    0.10
    uttle
    0.10
    inho
    0.10
    hread
    0.10
    ively
    0.09
    ewart
    0.09
    otland
    0.09
    ëĮĵ
    0.09
    æĪijè¿ĺ
    0.09
    alley
    0.09
    Activations Density 0.015%

    No Known Activations