© 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. 4418
    Prev
    Next
    INDEX
    Explanations

    unique values or identifiers

    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
    oti
    -0.08
    sexual
    -0.08
    ÙĴ
    -0.08
    atted
    -0.08
    aceous
    -0.08
    ouse
    -0.08
    acos
    -0.08
    ëıĻ
    -0.08
    ä¹ĭ家
    -0.08
    osoph
    -0.07
    POSITIVE LOGITS
    orns
    0.12
    ely
    0.12
    itarian
    0.12
    è§ģ解
    0.11
    zers
    0.10
    ìłģìľ¼ë¡ľ
    0.10
    å¢ħ
    0.10
    íķĺê²Į
    0.10
    ities
    0.10
    ties
    0.09
    Activations Density 0.047%

    No Known Activations