© 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. 30258
    Prev
    Next
    INDEX
    Explanations

    numeric sequences with units or followed by words

    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
    æĦı大
    -0.10
    åĵŁ
    -0.09
    çĴ°
    -0.08
    è£ģ
    -0.08
    æĭ¾
    -0.08
    çݯ
    -0.08
    arness
    -0.08
    媳
    -0.08
    游åĩ»
    -0.07
    æ¤ħ
    -0.07
    POSITIVE LOGITS
    FIG
    0.09
    tone
    0.09
    boys
    0.08
    XR
    0.08
     наÑĥÑĩ
    0.08
    BYTE
    0.08
    cratch
    0.08
    WORDS
    0.08
    NSError
    0.08
    ìļ©
    0.08
    Activations Density 1.911%

    No Known Activations