© 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. 33781
    Prev
    Next
    INDEX
    Explanations

    will/has/may/would followed by verb

    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
    åIJĦæł·
    -0.09
    åħ±åIJĮåıijå±ķ
    -0.09
    yers
    -0.08
    å§ĭ
    -0.08
    esses
    -0.08
    å¥ļ
    -0.08
    åĬłåħ¥äºĨ
    -0.08
    çŁ³æ¦´
    -0.08
    dsl
    -0.08
    è§īå¾Ĺèĩªå·±
    -0.08
    POSITIVE LOGITS
    æĦıåij³çĿĢ
    0.12
    éļ¾æĢª
    0.11
    åıijçĶŁåľ¨
    0.11
     ÑģлÑĥÑĩ
    0.10
     happened
    0.10
    çĹĽ
    0.10
    éŵ
    0.09
    æĹĹ
    0.09
    iek
    0.09
     happening
    0.08
    Activations Density 0.044%

    No Known Activations