© 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. 30897
    Prev
    Next
    INDEX
    Explanations

    How to handle

    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
    åĢį
    -0.10
    æĺ§
    -0.10
    .Utility
    -0.09
    ForeignKey
    -0.09
    åĬŁå¤«
    -0.09
    éŀĭ
    -0.08
    _phrase
    -0.08
    æ±Łå±±
    -0.08
    venge
    -0.08
    Firebase
    -0.08
    POSITIVE LOGITS
    åħ³åĪĩ
    0.09
     ак
    0.09
    æ£ĺ
    0.09
    çļĦåIJĦç§į
    0.09
    åºŀ大
    0.08
    é«ĺ级
    0.08
     á»ķ
    0.08
     вопÑĢоÑģÑĭ
    0.08
    æ¬
    0.08
    çIJIJ
    0.08
    Activations Density 0.092%

    No Known Activations