Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 84573
    Prev
    Next
    INDEX
    Explanations

    Space missions

    np_max-act · gemini-2.0-flash

    The neuron detects “As a …” style role-play or persona-setting statements (e.g. “As a space colonist on Mars…”).

    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
     grill
    -0.07
    FA
    -0.07
    �
    -0.07
    139
    -0.06
     århus
    -0.06
     BF
    -0.06
     Lisp
    -0.06
     pursuits
    -0.06
     pudding
    -0.06
     الموس
    -0.06
    POSITIVE LOGITS
    ें↵
    0.07
     \'
    0.07
    must
    0.06
     methodName
    0.06
    rial
    0.06
    Variables
    0.06
    .');↵↵
    0.06
     forks
    0.06
    ..........
    0.06
    	change
    0.06
    Activations Density 0.037%

    No Known Activations