© 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. 13628
    Prev
    Next
    INDEX
    Explanations

    legitimate

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of the word “legitimate” (or its stems) in the text.

    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
     hoa
    -0.08
    १
    -0.07
     дія
    -0.07
     adherence
    -0.07
     intptr
    -0.06
    8
    -0.06
    िवर
    -0.06
    [N
    -0.06
    اسر
    -0.06
     tratamiento
    -0.06
    POSITIVE LOGITS
     legitimate
    0.13
     legitimacy
    0.09
     legit
    0.09
    itimate
    0.08
    nett
    0.07
    'name
    0.07
     Leg
    0.07
    лем
    0.07
    emek
    0.07
     legitim
    0.07
    Activations Density 0.003%

    No Known Activations