© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 8989
    Prev
    Next
    INDEX
    Explanations

    I

    np_max-act · gemini-2.0-flash

    This neuron activates on tokens from the hidden system or persona‐setup instructions (e.g. “This,” “He,” “His,” numerical metadata) rather than on actual conversation content.

    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
    SHIFT
    -0.06
    pri
    -0.06
    _ALT
    -0.06
    terrorism
    -0.06
    coffee
    -0.06
    _TASK
    -0.06
     Hein
    -0.06
     Libertarian
    -0.06
    cedes
    -0.06
     counterparts
    -0.06
    POSITIVE LOGITS
     NSError
    0.08
    (MenuItem
    0.07
    努力
    0.07
    _INFINITY
    0.07
    .gc
    0.07
     toss
    0.07
    coder
    0.07
     |--------------------------------------------------------------------------↵
    0.07
     символ
    0.06
    ักษณ
    0.06
    Activations Density 0.053%

    No Known Activations