© 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. 8301
    Prev
    Next
    INDEX
    Explanations

    cold conditions or states

    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
    åĨħåľ¨
    -0.11
    èĵį
    -0.11
    aler
    -0.10
    Sugar
    -0.10
    ALLE
    -0.09
    NonNull
    -0.09
    ERIC
    -0.09
    髯
    -0.09
    çĮĽ
    -0.09
    ALER
    -0.09
    POSITIVE LOGITS
    åĩī
    0.14
    åĨ·
    0.13
    Cool
    0.13
    åĨ·éĿĻ
    0.13
     cool
    0.13
     Ñħолод
    0.12
     cold
    0.11
    cold
    0.11
    cool
    0.11
    åĨ·æ°´
    0.11
    Activations Density 0.080%

    No Known Activations