© 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. 33212
    Prev
    Next
    INDEX
    Explanations

    difference between

    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
    åģļçļĦ
    -0.08
    赶ä¸Ĭ
    -0.08
    æ¹
    -0.08
    éĺĶ
    -0.08
     sobie
    -0.08
    BOR
    -0.08
    ourt
    -0.08
    holm
    -0.08
    ATORY
    -0.08
    Detach
    -0.07
    POSITIVE LOGITS
    çļĦåĮºåĪ«
    0.25
    åĮºåĪ«
    0.25
    å·®åĪ«
    0.23
     difference
    0.21
    åĮºåĪĨ
    0.20
     distinction
    0.19
    Difference
    0.18
    difference
    0.17
    å·®å¼Ĥ
    0.17
     Difference
    0.16
    Activations Density 0.228%

    No Known Activations