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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 92186
    Prev
    Next
    INDEX
    Explanations

    order

    np_max-act · gemini-2.0-flash

    This neuron detects terms related to maintaining order or stability (e.g., words built on “ord–” or “stabil–”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    زيد
    -0.07
    icích
    -0.07
     情
    -0.06
    лений
    -0.06
     함수
    -0.06
    -release
    -0.06
    ソ
    -0.06
    ่ง
    -0.06
     разработ
    -0.06
     Suicide
    -0.06
    POSITIVE LOGITS
     orderly
    0.07
     indiv
    0.07
     sched
    0.07
    /Admin
    0.06
    協
    0.06
     Ả
    0.06
    Learn
    0.06
     agreed
    0.06
    PrototypeOf
    0.06
    ｗ
    0.06
    Activations Density 0.018%

    No Known Activations