© 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. 53981
    Prev
    Next
    INDEX
    Explanations

    known for / known as

    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
    emy
    -0.09
    æ·º
    -0.08
    åıijå±ķæľºéģĩ
    -0.08
    éĻĪåĪĹ
    -0.08
    ç³Ł
    -0.08
    'd
    -0.08
    æݧ
    -0.08
    çļĦ人çĶŁ
    -0.08
    enth
    -0.07
    æĪijè¿ĺæĺ¯
    -0.07
    POSITIVE LOGITS
     known
    0.12
     Known
    0.11
    以åħ¶
    0.11
    known
    0.11
     извеÑģÑĤ
    0.11
    æŁļ
    0.11
     bekannt
    0.11
    _known
    0.10
    -known
    0.10
    ointed
    0.10
    Activations Density 0.072%

    No Known Activations