© 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. 10228
    Prev
    Next
    INDEX
    Explanations

    intention to / of

    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
     Ñģобой
    -0.12
    ulence
    -0.12
    ings
    -0.09
    -toggler
    -0.09
    ture
    -0.09
    thest
    -0.09
    abra
    -0.09
    ÂŃing
    -0.09
    ler
    -0.09
    itecture
    -0.09
    POSITIVE LOGITS
    nement
    0.12
    ality
    0.11
    naire
    0.10
    ary
    0.10
    çĽİ
    0.09
    nelle
    0.09
    è°Ĭ
    0.09
    Ñĥд
    0.09
    Bindable
    0.09
    ate
    0.09
    Activations Density 0.006%

    No Known Activations