© 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. 9002
    Prev
    Next
    INDEX
    Explanations

    adverse reaction or response

    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
     yön
    -0.09
    绣é¢Ĩ
    -0.09
     bombs
    -0.09
    åħħ满äºĨ
    -0.08
    åħħ满
    -0.08
    æĸ©
    -0.08
     Chronicle
    -0.08
     bomb
    -0.08
    UTION
    -0.08
    ç»ķ
    -0.08
    POSITIVE LOGITS
    åıįåºĶ
    0.16
     reaction
    0.14
    åıįæĩī
    0.13
     response
    0.13
     reactions
    0.12
     ÑĢеак
    0.12
    åıįåĵį
    0.11
     responses
    0.11
    强çĥĪçļĦ
    0.11
    _responses
    0.11
    Activations Density 0.118%

    No Known Activations