© 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. 47892
    Prev
    Next
    INDEX
    Explanations

    best

    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
    å½¹
    -0.11
    çĸ¾çĹħçļĦ
    -0.09
    æľīä¸ĢçĤ¹
    -0.08
    ember
    -0.08
    omes
    -0.08
     studio
    -0.08
    è½½
    -0.08
    ç¦Ħ
    -0.08
    acs
    -0.08
    IGIN
    -0.08
    POSITIVE LOGITS
    æľĢ好çļĦ
    0.20
     best
    0.17
    æľĢä½³
    0.15
    best
    0.15
    Best
    0.14
     Best
    0.14
     beste
    0.14
     лÑĥÑĩÑĪиÑħ
    0.14
    (best
    0.14
     brightest
    0.14
    Activations Density 0.098%

    No Known Activations