© 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. 62320
    Prev
    Next
    INDEX
    Explanations

    bra

    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
    庹
    -0.10
    ubble
    -0.10
    ings
    -0.10
    æĥ³åĬŀæ³ķ
    -0.09
    ç²Ĵ
    -0.09
    gran
    -0.09
    åħĪéĶĭ
    -0.09
    aving
    -0.09
    åįļè§Ī
    -0.08
    æIJª
    -0.08
    POSITIVE LOGITS
    _ACTIV
    0.10
    Ñħи
    0.10
    æĢ¯
    0.09
    rame
    0.09
    eba
    0.08
    ðŁı»
    0.08
    è©
    0.08
    ead
    0.08
     вопÑĢоÑģ
    0.08
    ếp
    0.08
    Activations Density 0.006%

    No Known Activations