© 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. 29876
    Prev
    Next
    INDEX
    Explanations

    back in [year/time]

    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
    åIJİç»Ń
    -0.10
    ä»İèĢĮ使
    -0.09
    rog
    -0.09
    å½ĵæŶçļĦ
    -0.09
    rew
    -0.09
     Byron
    -0.09
    æĿ¥çĶµ
    -0.09
    åĮĻ
    -0.09
    使ä¹ĭ
    -0.09
    ä¸ĵ线
    -0.09
    POSITIVE LOGITS
    wards
    0.12
    -fashioned
    0.09
    èĻij
    0.09
    OrDefault
    0.09
    ward
    0.09
     awhile
    0.09
    ä¹İ
    0.09
    WARDS
    0.08
    WARD
    0.08
    /current
    0.08
    Activations Density 0.011%

    No Known Activations