© 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. 7285
    Prev
    Next
    INDEX
    Explanations

    by the noun

    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
    ника
    -0.09
    otts
    -0.09
    è¨ĢãģĨ
    -0.08
    ElementException
    -0.08
    DDL
    -0.08
    çĭ¬
    -0.08
    顺
    -0.08
    opa
    -0.08
    apes
    -0.08
    被称为
    -0.08
    POSITIVE LOGITS
    åĽłä¸ºå®ĥ
    0.12
    ipy
    0.11
    ä¸įäºĪ
    0.10
    çľĭå¾ħ
    0.09
    èµ°ä¸Ĭ
    0.09
    ä¸įæ¸ħ
    0.09
    ously
    0.09
    å¼Ģåıij建设
    0.08
    èį£èªī称åı·
    0.08
    éĥ¨åĪĨåľ°åĮº
    0.08
    Activations Density 0.192%

    No Known Activations