© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 41582
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron fires on the punctuation delimiters (commas, periods, semicolons) that separate names and titles in the case caption/header.

    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
    >User
    -0.07
    (parent
    -0.06
    Jamie
    -0.06
    Brian
    -0.06
    _extended
    -0.06
    न
    -0.06
     jeszcze
    -0.06
    采用
    -0.06
    instruction
    -0.06
     Dare
    -0.06
    POSITIVE LOGITS
    AdapterManager
    0.07
    '%(
    0.06
     greed
    0.06
    .INTEGER
    0.06
     عليها
    0.06
    .BOLD
    0.06
    _CODES
    0.06
     mặc
    0.06
     огром
    0.06
     COURT
    0.06
    Activations Density 0.017%

    No Known Activations