© 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. 8319
    Prev
    Next
    INDEX
    Explanations

    right click menu

    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
    çīµ
    -0.09
    apes
    -0.09
    çľ·
    -0.09
    rame
    -0.08
    vale
    -0.08
    łéĻ¤
    -0.08
    pairs
    -0.08
    æ´½
    -0.08
    åĿĹ
    -0.08
    ç´¯
    -0.08
    POSITIVE LOGITS
    ç¿Ĭ
    0.09
     context
    0.09
    èıľåįķ
    0.09
    æĬ¤
    0.09
    CONTEXT
    0.08
    女æĺŁ
    0.08
    .ToolTip
    0.08
    åħ³èģĶ交æĺĵ
    0.08
    诱
    0.08
    stri
    0.08
    Activations Density 0.040%

    No Known Activations