© 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. 19662
    Prev
    Next
    INDEX
    Explanations

    help or provide

    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
    _DX
    -0.09
    iske
    -0.09
    icket
    -0.09
    ayne
    -0.09
    ilst
    -0.08
    缸éĤ»
    -0.08
     gettext
    -0.08
    ecn
    -0.08
    CanBeConverted
    -0.08
    çĥŁçģ«
    -0.08
    POSITIVE LOGITS
    çݰ代åĮĸ
    0.10
    æİĩ
    0.09
    çļĦ积æŀģæĢ§
    0.09
    çļĦè¿ĩç¨ĭä¸Ń
    0.08
    已被
    0.08
    éļ¾å¿ĺ
    0.08
    éĤ£ä»½
    0.08
    ä¸Ģ次次
    0.08
    æĶ¯éĥ¨
    0.08
     overall
    0.08
    Activations Density 0.046%

    No Known Activations