© 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. 126168
    Prev
    Next
    INDEX
    Explanations

    Wu

    np_max-act · gemini-2.0-flash

    This neuron detects placeholder participant identifiers of the form “NAME_<number>.”

    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
     shipments
    -0.08
    _aes
    -0.07
    iez
    -0.07
    dv
    -0.07
    ωτερ
    -0.06
    стве
    -0.06
    )');↵
    -0.06
    _funcs
    -0.06
     Criterion
    -0.06
    xima
    -0.06
    POSITIVE LOGITS
    卻
    0.06
     国家
    0.06
    TableModel
    0.06
    】,
    0.06
    ()].
    0.06
     enable
    0.06
    {/
    0.06
     PO
    0.06
    QUEST
    0.06
     compuls
    0.06
    Activations Density 0.003%

    No Known Activations