© 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. 6709
    Prev
    Next
    INDEX
    Explanations

    hal, halo, and halide

    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
    ercial
    -0.10
    eer
    -0.10
    ä»Ģä¹Īåij¢
    -0.09
    dale
    -0.09
    liness
    -0.09
    æį¢åıĸ
    -0.09
    errupt
    -0.09
    ingen
    -0.09
    è¿Ļæł·ä¸Ģ
    -0.08
    elta
    -0.08
    POSITIVE LOGITS
    ä»İ严治åħļ
    0.10
    tern
    0.10
    ëģĶ
    0.09
    opsis
    0.09
    tered
    0.09
    è¿
    0.09
    æİī
    0.09
    weg
    0.09
     Ñģобой
    0.09
    	TokenName
    0.08
    Activations Density 0.026%

    No Known Activations