© 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. 56745
    Prev
    Next
    INDEX
    Explanations

    ÂĴ followed by 're', 't', 's', 'comments'

    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
    глÑı
    -0.09
    åħ»
    -0.08
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    -0.08
     Canary
    -0.08
    letters
    -0.08
    a
    -0.08
    Gar
    -0.07
    çĤ¹åĦ¿
    -0.07
    .refs
    -0.07
    t
    -0.07
    POSITIVE LOGITS
    ments
    0.12
    atives
    0.10
    olated
    0.10
    ories
    0.10
    mania
    0.10
    NewLabel
    0.10
    arians
    0.09
    ative
    0.09
     trous
    0.09
    nen
    0.09
    Activations Density 0.020%

    No Known Activations