© 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. 2598
    Prev
    Next
    INDEX
    Explanations

    backwards compatibility

    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
    æīĭå¥Ĺ
    -0.10
    çķĮ
    -0.10
    readonly
    -0.09
    æīĵ好
    -0.08
    éļ¨èijĹ
    -0.08
     PUT
    -0.08
    èģĶç½ij
    -0.08
    èķ²
    -0.08
     Flour
    -0.08
     Bermuda
    -0.08
    POSITIVE LOGITS
    åħ¼å®¹
    0.18
    legacy
    0.15
     legacy
    0.15
    æŧ
    0.14
     compatibility
    0.14
     compat
    0.13
    compat
    0.13
     Compatibility
    0.12
    Compatibility
    0.12
    èĪĬ
    0.12
    Activations Density 0.179%

    No Known Activations