© 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. 57755
    Prev
    Next
    INDEX
    Explanations

    proof or evidence

    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
    ewise
    -0.10
    ="{!!
    -0.09
     мн
    -0.09
     Zoo
    -0.08
    纡
    -0.08
     Mine
    -0.08
     Zo
    -0.08
    simd
    -0.08
     Came
    -0.08
     Inspir
    -0.08
    POSITIVE LOGITS
    è¯ģæį®
    0.17
     evidence
    0.14
    èŃī
    0.13
    åıĸè¯ģ
    0.12
    Evidence
    0.11
    åĩŃè¯ģ
    0.11
     Evidence
    0.10
    è¯ģæĺİ
    0.10
    å¤ĩæ¡Ī
    0.10
    éĻŁ
    0.10
    Activations Density 0.040%

    No Known Activations