© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38050
    Prev
    Next
    INDEX
    Explanations

    reward/punish consequences

    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
    çĶĵ
    -0.10
    éĻ¢éĻ¢å£«
    -0.09
     Hod
    -0.09
    è¾ŀ
    -0.09
    åī¯
    -0.08
    æŁ©
    -0.08
    Forge
    -0.08
    å·¥åİĤ
    -0.08
    è¾Ń
    -0.08
    ç»Łè®¡æķ°æį®
    -0.08
    POSITIVE LOGITS
    å¥ĸåĬ±
    0.15
     rewarded
    0.14
     reward
    0.14
     Reward
    0.14
     punished
    0.12
     rewarding
    0.12
    reward
    0.11
    æĥ©ç½ļ
    0.11
     punish
    0.11
     rewards
    0.11
    Activations Density 0.022%

    No Known Activations