© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8560
    Prev
    Next
    INDEX
    Explanations

    than other comparisons

    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
    ARING
    -0.10
    leta
    -0.10
    sWith
    -0.10
    å¨ĺå¨ĺ
    -0.10
    ported
    -0.09
    ily
    -0.09
    æŃ¤äºº
    -0.09
    å¨ĺ家
    -0.09
    robat
    -0.09
    /operator
    -0.09
    POSITIVE LOGITS
    éĺĹ
    0.10
    atology
    0.09
    ãģ£ãģįãĤĬ
    0.09
    /sm
    0.09
    åħ¶ä»ĸ
    0.09
     existing
    0.09
     others
    0.09
    SOR
    0.09
    than
    0.09
    iento
    0.08
    Activations Density 0.079%

    No Known Activations