© 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. 51228
    Prev
    Next
    INDEX
    Explanations

    quotation marks and non-Latin scripts

    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
    åij¼
    -0.10
     Outcome
    -0.09
    åĪº
    -0.08
    Outcome
    -0.08
    åħħ
    -0.08
    trail
    -0.08
    ega
    -0.08
    æ©ĺ
    -0.08
    aning
    -0.07
    itzer
    -0.07
    POSITIVE LOGITS
     ÐłÐ¾ÑģÑģ
    0.11
    åħ¶å®ĥéĹ®é¢ĺ
    0.10
    åħ¶ä»ĸçļĦ
    0.09
    çĤ«èĢĢ
    0.09
    unker
    0.09
    çĺ¾
    0.09
    对ä¸įèµ·
    0.09
    åĭģ
    0.09
    ilit
    0.09
    å¦Ħ
    0.09
    Activations Density 0.015%

    No Known Activations