© 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. 50511
    Prev
    Next
    INDEX
    Explanations

    drag and drop

    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
    owing
    -0.12
    hood
    -0.09
    ãĤº
    -0.09
    волÑĮ
    -0.09
    roma
    -0.09
    Gravity
    -0.09
    볨
    -0.09
    æĬµ
    -0.09
    TING
    -0.09
    аз
    -0.09
    POSITIVE LOGITS
    gable
    0.11
    uet
    0.11
    åĩºå¤Ħ
    0.10
    apult
    0.10
    ues
    0.10
    åıįæŃ£
    0.09
    gings
    0.09
    arra
    0.09
    uese
    0.09
    èĽĶ
    0.09
    Activations Density 0.009%

    No Known Activations