© 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. 48105
    Prev
    Next
    INDEX
    Explanations

    cont prefixes

    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
    -0.17
    rd
    -0.11
    -quarters
    -0.10
    ä»Ģä¹ĪåİŁåĽł
    -0.09
    otic
    -0.09
    游åĩ»éĺŁ
    -0.09
    nda
    -0.09
    horse
    -0.09
    gone
    -0.09
    éħIJ
    -0.09
    POSITIVE LOGITS
    éĻIJãĤĬ
    0.13
    ents
    0.11
    aneous
    0.10
    -thirds
    0.10
    orary
    0.10
    iveness
    0.10
    ively
    0.10
    iguous
    0.10
    uous
    0.10
    uously
    0.10
    Activations Density 0.035%

    No Known Activations