© 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. 49578
    Prev
    Next
    INDEX
    Explanations

    repository section or structure

    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
    代表大ä¼ļ
    -0.10
    tober
    -0.10
    heet
    -0.09
    most
    -0.09
    obb
    -0.09
    ippers
    -0.09
    uates
    -0.08
    à¹Ĩ
    -0.08
    inals
    -0.08
    uator
    -0.08
    POSITIVE LOGITS
    ection
    0.11
    åĨħçļĦ
    0.10
    itories
    0.10
    OrUpdate
    0.09
    çĪ±å°Ķåħ°
    0.09
    窿
    0.09
    å¢ħ
    0.09
    wers
    0.08
    rupt
    0.08
    cats
    0.08
    Activations Density 0.039%

    No Known Activations