© 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. 23028
    Prev
    Next
    INDEX
    Explanations

    typographical errors

    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
    inous
    -0.09
    oft
    -0.09
    bn
    -0.08
    èĬŁ
    -0.08
    MediaType
    -0.08
    lings
    -0.08
    ability
    -0.08
    maids
    -0.08
     TU
    -0.08
    «
    -0.08
    POSITIVE LOGITS
    ãģĹãĤĩãģĨ
    0.10
    MBER
    0.09
    ersen
    0.09
    éĤ»å±ħ
    0.08
    ISTIC
    0.08
    getReference
    0.08
    ÄĻ
    0.08
     erst
    0.08
    iscal
    0.08
    uration
    0.08
    Activations Density 0.013%

    No Known Activations