© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20061
    Prev
    Next
    INDEX
    Explanations

    rules

    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
    omes
    -0.11
    InParameter
    -0.10
    çľ¼åħī
    -0.09
    olec
    -0.09
    åĽŀæĥ³
    -0.09
    oppers
    -0.09
    ä¸ºæľŁ
    -0.09
    æ¢ħèĬ±
    -0.09
    obuf
    -0.09
    reich
    -0.09
    POSITIVE LOGITS
     Rules
    0.13
     rules
    0.12
     RULE
    0.11
    è§ĦåĪĻ
    0.11
    .Rule
    0.10
    (rules
    0.10
    rules
    0.10
     thumb
    0.10
    ets
    0.10
    (rule
    0.09
    Activations Density 0.055%

    No Known Activations