© 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. 50837
    Prev
    Next
    INDEX
    Explanations

    problem or issue

    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
    uces
    -0.10
    è²Į
    -0.10
    æĤ²åī§
    -0.09
    åIJĦè¡Į
    -0.09
    helm
    -0.09
    ä»ĩ
    -0.09
    åĩłåIJį
    -0.09
    _BOUND
    -0.09
    èĭĹ
    -0.09
    å°½
    -0.08
    POSITIVE LOGITS
     concern
    0.15
    çļĦéĹ®é¢ĺ
    0.13
    æĭħå¿ĥ
    0.13
     worry
    0.12
    åĽ°æī°
    0.12
     concerns
    0.12
     issues
    0.12
     issue
    0.12
     worries
    0.12
     Concern
    0.11
    Activations Density 0.077%

    No Known Activations