© 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. 55887
    Prev
    Next
    INDEX
    Explanations

    revealing information

    np_max-act · gemini-2.0-flash

    This neuron activates on words that appear in disclaimers about unverified information, i.e. in phrases signaling that the content isn’t confirmed.

    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
    宝
    -0.07
    (StringUtils
    -0.06
    �
    -0.06
    _corner
    -0.06
    -ip
    -0.06
    ]")↵
    -0.06
    ?):
    -0.06
    istine
    -0.06
    Liquid
    -0.06
     loa
    -0.06
    POSITIVE LOGITS
    executable
    0.06
     prolong
    0.06
     prompts
    0.06
    .start
    0.06
     companyName
    0.06
    .ConnectionStrings
    0.06
     prompted
    0.06
     Begin
    0.06
    -report
    0.06
     Toyota
    0.06
    Activations Density 0.001%

    No Known Activations