© 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. 61478
    Prev
    Next
    INDEX
    Explanations

    field and off-field behaviour

    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
    ewe
    -0.11
    .Writer
    -0.10
    å¡ŀ
    -0.10
    ritel
    -0.10
    .EMPTY
    -0.10
     ÅĽc
    -0.09
    æµĴ
    -0.09
    ún
    -0.09
    lse
    -0.09
    éŨåı£
    -0.09
    POSITIVE LOGITS
    ignum
    0.10
    åĸľæ¬¢åIJĥ
    0.09
    表çĻ½
    0.09
    æIJŀ
    0.08
    份
    0.08
    éķIJ
    0.08
    缴è¨Ģ
    0.08
    åIJĥçļĦ
    0.08
    åģļ人
    0.08
    èĩªè§ī
    0.08
    Activations Density 0.122%

    No Known Activations