© 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. 9923
    Prev
    Next
    INDEX
    Explanations

    two-factor

    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
    åĶ¿
    -0.13
    æį½
    -0.08
     admon
    -0.08
    \_
    -0.08
    è¯Ľ
    -0.07
     bisc
    -0.07
    Č
    -0.07
    >b
    -0.07
    è¿ĺæľīä»Ģä¹Ī
    -0.07
    ordo
    -0.06
    POSITIVE LOGITS
    åħ¬ç½ijå®īå¤ĩ
    0.10
    oner
    0.09
    odore
    0.09
    å¸ļ
    0.09
    åĴĮåľ°åĮº
    0.09
    çİĭçīĮ
    0.08
    ç½ij约车
    0.08
    èĭ±çī¹
    0.08
    âĨĵ
    0.08
    uffer
    0.08
    Activations Density 0.726%

    No Known Activations