© 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. 36897
    Prev
    Next
    INDEX
    Explanations

    word preceding 'you' or 'all'

    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
    assen
    -0.09
    oting
    -0.09
    dap
    -0.09
    gree
    -0.08
    oters
    -0.08
    _pa
    -0.08
    UTDOWN
    -0.08
    éĤ£ç§į
    -0.08
    åĬ¨åijĺ
    -0.08
    utdown
    -0.08
    POSITIVE LOGITS
    çļĦæīĢæľī
    0.20
    æīĢæľī
    0.18
     all
    0.18
     вÑģе
    0.17
    æīĢæľīçļĦ
    0.16
    ãģĻãģ¹ãģ¦
    0.16
    åħ¨éĥ¨
    0.16
     모ëijIJ
    0.15
     вÑģеÑħ
    0.15
     вÑģÑİ
    0.13
    Activations Density 0.345%

    No Known Activations