© 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. 91590
    Prev
    Next
    INDEX
    Explanations

    feelings and emotions

    np_max-act · gemini-2.0-flash

    The neuron fires on empathetic or reassuring words and short phrases (e.g. “normal,” “important,” “understand,” “can,” “time”) that occur in supportive, consolation‐style responses.

    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
     Perhaps
    -0.07
     oli
    -0.07
    PushMatrix
    -0.06
     jugar
    -0.06
    .Threading
    -0.06
     Make
    -0.06
    potential
    -0.06
    xml
    -0.06
     explanation
    -0.06
     todo
    -0.06
    POSITIVE LOGITS
    STRU
    0.06
    0.06
    hana
    0.06
     dinosaurs
    0.06
    setError
    0.06
    .appspot
    0.06
    lama
    0.06
     calle
    0.06
    탁
    0.06
    TTY
    0.06
    Activations Density 0.023%

    No Known Activations