© 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. 62352
    Prev
    Next
    INDEX
    Explanations

    horizon

    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
    sense
    -0.10
    out
    -0.10
    itative
    -0.09
    engu
    -0.09
    ish
    -0.08
    деÑģÑıÑĤ
    -0.08
    erv
    -0.08
    eral
    -0.08
    erness
    -0.08
    uman
    -0.08
    POSITIVE LOGITS
    ä¸įæ¸ħ
    0.10
    广éĺĶ
    0.10
    -reaching
    0.10
    éĩįéĩį
    0.10
    raph
    0.10
    áln
    0.09
    nement
    0.09
    åĨħçļĦ
    0.09
    æľĽåİ»
    0.09
    尽头
    0.09
    Activations Density 0.006%

    No Known Activations