© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51743
    Prev
    Next
    INDEX
    Explanations

    cave related words

    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
    ÑĤе
    -0.10
    à¹Ģà¸Ńà¸ĩ
    -0.10
     rowspan
    -0.09
    ยà¸ģ
    -0.09
    mitter
    -0.09
    alie
    -0.09
    LEV
    -0.09
    owanie
    -0.09
    -demand
    -0.08
    è¾Ĩ
    -0.08
    POSITIVE LOGITS
    ç©´
    0.12
    窿
    0.12
    OrUpdate
    0.10
    æĤī
    0.10
    å¹½
    0.10
     Progressive
    0.09
    æ·±å¤Ħ
    0.09
    IRD
    0.09
    iband
    0.09
    ãģ«åħ¥ãĤĭ
    0.09
    Activations Density 0.023%

    No Known Activations