© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 53015
    Prev
    Next
    INDEX
    Explanations

    fred

    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
    ally
    -0.12
    ãģ¹ãģį
    -0.12
    edList
    -0.11
    ¡
    -0.11
    ãĥ£
    -0.11
    ij
    -0.10
     principalTable
    -0.10
    ality
    -0.10
    管å§Ķä¼ļ
    -0.10
    teenth
    -0.10
    POSITIVE LOGITS
    illet
    0.10
    ious
    0.09
    奶奶
    0.09
    ðŁı»
    0.08
    ium
    0.08
    åijĺ
    0.08
    usz
    0.08
    rence
    0.08
    uzzi
    0.08
    iras
    0.08
    Activations Density 0.010%

    No Known Activations