© 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. 92102
    Prev
    Next
    INDEX
    Explanations

    called

    np_max-act · gemini-2.0-flash

    This neuron detects the use of sarcastic or dismissive qualifiers (e.g. “so-called,” “pretend,” scare-quotes) that signal skepticism or irony.

    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
    -make
    -0.06
     wildly
    -0.06
    adel
    -0.06
    rections
    -0.06
     qualquer
    -0.06
     预
    -0.06
     Flores
    -0.06
     متفاوت
    -0.06
    asers
    -0.06
    -right
    -0.06
    POSITIVE LOGITS
     ostensibly
    0.07
    .Lerp
    0.07
     Indexed
    0.07
     نف
    0.07
    ToolBar
    0.07
    @GeneratedValue
    0.07
     supposedly
    0.07
    .LOG
    0.07
     allegedly
    0.07
    .uuid
    0.06
    Activations Density 0.023%

    No Known Activations