© 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. 68880
    Prev
    Next
    INDEX
    Explanations

    reflected

    np_max-act · gemini-2.0-flash

    This neuron detects descriptive past-participles (–ed forms) used in formal or academic contexts (e.g. “embodied,” “reflected,” “represented,” “exemplified”).

    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
     undergrad
    -0.07
    Href
    -0.07
    werp
    -0.06
    	RTCT
    -0.06
     undergraduate
    -0.06
    But
    -0.06
    MQ
    -0.06
    enschaft
    -0.06
     localVar
    -0.06
    _type
    -0.06
    POSITIVE LOGITS
     کلی
    0.07
     شی
    0.07
     política
    0.07
    ciler
    0.07
     kelim
    0.06
     прием
    0.06
    RCT
    0.06
    ACHI
    0.06
    >r
    0.06
     각각
    0.06
    Activations Density 0.040%

    No Known Activations