© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 6562
    Prev
    Next
    INDEX
    Explanations

    roman numerals and numbers

    np_max-act · gemini-2.0-flash

    This neuron detects section or tutorial headings that include “Part” followed by a Roman numeral or digit (e.g. “Part II,” “III,” “2”).

    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
     playable
    -0.06
     teachers
    -0.06
    介绍
    -0.06
    eland
    -0.06
     means
    -0.06
    aphrag
    -0.06
     lö
    -0.06
    	Server
    -0.06
     disastr
    -0.06
     موجب
    -0.06
    POSITIVE LOGITS
     prostit
    0.08
     phil
    0.07
    옥
    0.07
    icut
    0.06
    osten
    0.06
    _APB
    0.06
    โล
    0.06
     Cary
    0.06
    민
    0.06
    etest
    0.06
    Activations Density 0.021%

    No Known Activations