© 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. 11627
    Prev
    Next
    INDEX
    Explanations

    refueling, refilling, or recharging

    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
     Lun
    -0.08
    itia
    -0.08
    _RD
    -0.08
    éĺ¶æ®µæĢ§
    -0.08
    ective
    -0.08
    é«ĺå±±
    -0.08
    iltr
    -0.08
    .SingleOrDefault
    -0.08
     exhaust
    -0.08
     Ward
    -0.08
    POSITIVE LOGITS
     replen
    0.13
     refill
    0.11
    åħħ足
    0.11
    aug
    0.10
    åĤ¨å¤ĩ
    0.09
    serter
    0.09
    è¡¥åħħ
    0.09
    hé
    0.09
    odore
    0.09
    æ·»
    0.09
    Activations Density 0.059%

    No Known Activations