© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 89535
    Prev
    Next
    INDEX
    Explanations

    Generic articles

    np_max-act · gemini-2.0-flash

    The neuron fires most strongly on emotionally charged or evaluative content words—particularly subjective adjectives and adverbs (e.g. “heart-wrenching,” “beautiful,” “literally”) that convey personal sentiment or emphasis.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    elements related to personal experiences and emotional responses.

    oai_token-act-pair · gpt-4o-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
     moż
    -0.07
    лок
    -0.07
    ур
    -0.07
     fanatic
    -0.07
    illator
    -0.06
     іде
    -0.06
    iếp
    -0.06
     brewers
    -0.06
     giy
    -0.06
    .listFiles
    -0.06
    POSITIVE LOGITS
    );\
    0.07
     Giải
    0.06
    dration
    0.06
    Thin
    0.06
    woff
    0.06
     Akt
    0.06
     Hoff
    0.06
     Reve
    0.06
    -select
    0.06
    ské
    0.06
    Activations Density 0.180%

    No Known Activations