© 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. 6448
    Prev
    Next
    INDEX
    Explanations

    ethical dilemmas

    np_max-act · gemini-2.0-flash

    This neuron fires on language signaling an inner, first-person ethical or emotional self-reflection—terms like “inner monologue,” “ethical,” “turmoil,” and related self-referential cues.

    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
     �
    -0.07
    _cut
    -0.07
     Azerbai
    -0.07
     Vern
    -0.06
     claim
    -0.06
    LANGUAGE
    -0.06
     Leigh
    -0.06
     itibaren
    -0.06
     allowances
    -0.06
     ダ
    -0.06
    POSITIVE LOGITS
    _alive
    0.07
     literals
    0.07
     SOCK
    0.06
    .Observable
    0.06
    _EXT
    0.06
     Adoption
    0.06
    _OCCURRED
    0.06
     관리
    0.06
     mary
    0.06
    jsx
    0.06
    Activations Density 0.012%

    No Known Activations