© 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. 11157
    Prev
    Next
    INDEX
    Explanations

    reclining, reclined, reclaim

    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
    aina
    -0.11
    quine
    -0.10
    åĭĥåĭĥ
    -0.10
    ãĤ£
    -0.09
    rite
    -0.09
    ako
    -0.09
    ว
    -0.09
    rong
    -0.09
    matic
    -0.09
     NOAA
    -0.08
    POSITIVE LOGITS
    ively
    0.12
    ../../../
    0.09
    çĶļ
    0.09
    ish
    0.09
    ¹
    0.08
    cmc
    0.08
    omedical
    0.08
    è·¯çģ¯
    0.08
    Codec
    0.08
    urring
    0.08
    Activations Density 0.027%

    No Known Activations