© 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. 38420
    Prev
    Next
    INDEX
    Explanations

    foreign 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
    è¿Ļæł·ä¸ĢæĿ¥
    -0.10
    wards
    -0.09
    zu
    -0.09
    zel
    -0.08
    è¿Ļæł·æīįèĥ½
    -0.08
    bra
    -0.08
    u
    -0.08
    w
    -0.08
    enga
    -0.08
     proverb
    -0.08
    POSITIVE LOGITS
    åĬ¿å¿ħ
    0.11
    èİ«è¿ĩäºİ
    0.10
    åĪĻæĺ¯
    0.10
    å°±ä¼ļ
    0.10
    èĩªçĦ¶
    0.10
    便
    0.10
    åīĩ
    0.10
    åį´åıijçİ°
    0.10
    eping
    0.10
    ieur
    0.10
    Activations Density 0.390%

    No Known Activations