© 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. 62972
    Prev
    Next
    INDEX
    Explanations

    endorse or ENDOR experiment

    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
    istics
    -0.09
    -Fi
    -0.09
    åĺ±
    -0.09
    apult
    -0.09
    itations
    -0.08
    åĪĨãģij
    -0.08
    tempt
    -0.08
    boats
    -0.08
    acks
    -0.08
    iska
    -0.08
    POSITIVE LOGITS
    æľŁåĨħ
    0.10
    amble
    0.09
    able
    0.09
    agger
    0.09
    zej
    0.09
    eds
    0.09
    IBLE
    0.09
    æĪijåİ»
    0.09
    rection
    0.09
     Isle
    0.09
    Activations Density 0.014%

    No Known Activations