© 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. 3791
    Prev
    Next
    INDEX
    Explanations

    reciprocal altruism, retaliatory actions

    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
     gar
    -0.09
    owy
    -0.09
    åĽ¤
    -0.09
    æľīå¾ħ
    -0.09
     stir
    -0.09
    éŁ³åĵį
    -0.09
    çļĦä¿¡å¿ĥ
    -0.09
    éĥĬ
    -0.08
    è·³è·ĥ
    -0.08
    avan
    -0.08
    POSITIVE LOGITS
     reciprocal
    0.17
     recipro
    0.17
     recip
    0.16
    äºĴåĪ©
    0.14
    äºĴ
    0.14
    对æĸ¹
    0.12
    äºĴéĢļ
    0.11
    对å¾ħ
    0.11
    äºĴ缸
    0.11
    å°įæĸ¹
    0.11
    Activations Density 0.071%

    No Known Activations