© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20480
    Prev
    Next
    INDEX
    Explanations

    prep for

    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
    ality
    -0.11
    beck
    -0.09
    oha
    -0.09
    hood
    -0.09
    rette
    -0.08
    å§ij
    -0.08
    .partner
    -0.08
    åīįä¸įä¹ħ
    -0.08
    ments
    -0.08
    FA
    -0.08
    POSITIVE LOGITS
    åĹĵåŃIJ
    0.10
    åĨĻåŃĹ
    0.10
    رار
    0.09
    üp
    0.09
    sters
    0.09
    çijŃ
    0.09
    èĢģ人家
    0.09
    SA
    0.09
    inspection
    0.08
    塬
    0.08
    Activations Density 0.007%

    No Known Activations