© 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. 51220
    Prev
    Next
    INDEX
    Explanations

    bottle or bottles

    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
    溷
    -0.09
    atics
    -0.09
    uate
    -0.09
    ools
    -0.09
     carte
    -0.08
    éĶľ
    -0.08
    atic
    -0.08
    åĽ¢åijĺ
    -0.08
    å¸IJ
    -0.08
    åı£åij³
    -0.08
    POSITIVE LOGITS
    çĵ¶
    0.12
     bottle
    0.10
     bottles
    0.10
     Bottle
    0.09
    (s
    0.09
    é¦Ļæ°´
    0.09
    gger
    0.09
    éħĴä¸ļ
    0.08
    ÑĤÑĥÑĢ
    0.08
    ciÄħ
    0.08
    Activations Density 0.023%

    No Known Activations