© 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. 682
    Prev
    Next
    INDEX
    Explanations

    tables of numbers and letters

    np_max-act · gemini-2.0-flash

    The neuron activates on meta‐instruction text and system‐level directives (e.g. “You are the text completion model…”) rather than on content words.

    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
     remake
    -0.07
     sense
    -0.06
    descriptor
    -0.06
    _tl
    -0.06
     Constructor
    -0.06
     setDescription
    -0.06
     Notre
    -0.06
     Judges
    -0.06
    .Reader
    -0.06
     خور
    -0.06
    POSITIVE LOGITS
    룹
    0.07
    aying
    0.06
     Downtown
    0.06
    ραν
    0.06
    雲
    0.06
    这种
    0.06
    (decoded
    0.06
    ın
    0.06
    Keys
    0.06
     outage
    0.06
    Activations Density 0.012%

    No Known Activations