© 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. 49927
    Prev
    Next
    INDEX
    Explanations

    walking and mobility

    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
    éºĿ
    -0.11
     hazırl
    -0.09
    FG
    -0.09
    éļĻ
    -0.09
    imon
    -0.09
    gest
    -0.09
    éĻį温
    -0.08
    çĸı导
    -0.08
    çĶŁå§ľ
    -0.08
    íĤ
    -0.08
    POSITIVE LOGITS
    èµ°è·¯
    0.18
    è¡Įèµ°
    0.16
    walking
    0.12
     walking
    0.12
    æŃ¥è¡Į
    0.12
    Walking
    0.12
     Walking
    0.12
    楼梯
    0.11
    _walk
    0.11
     crawling
    0.10
    Activations Density 0.079%

    No Known Activations