© 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. 15-RESID-POST-AA
    11. 33765
    Prev
    Next
    INDEX
    Explanations

    plot synopses

    np_max-act · gemini-2.0-flash

    References to caring for or taking responsibility for someone's well‑being (phrases like "cares for her", "taking care of", "making sure she's healthy").

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     всем
    -0.07
    VO
    -0.07
    '}}
    -0.07
    UND
    -0.07
     majors
    -0.06
    arker
    -0.06
    -pe
    -0.06
     ministers
    -0.06
    -light
    -0.06
    629
    -0.06
    POSITIVE LOGITS
     υπηρε
    0.06
     druhý
    0.06
    xac
    0.06
     indexer
    0.06
    (patient
    0.06
    (mark
    0.06
    (pub
    0.06
    DebugEnabled
    0.06
    	EXPECT
    0.06
     SHA
    0.06
    Activations Density 0.627%

    No Known Activations