© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 52392
    Prev
    Next
    INDEX
    Explanations

    thing

    np_max-act · gemini-2.0-flash

    expressions of support for societal responsibilities.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on words and phrases that express moral judgment—especially “right,” “correct,” “moral(ly),” or “the right thing to do.”

    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
     Weak
    -0.07
     dimensions
    -0.07
    _|
    -0.06
     bias
    -0.06
     هزار
    -0.06
     mild
    -0.06
    औ
    -0.06
     GDP
    -0.06
     эффектив
    -0.06
    	buff
    -0.06
    POSITIVE LOGITS
    asyarakat
    0.07
     serta
    0.07
    ountries
    0.06
    �
    0.06
    .Save
    0.06
    noDB
    0.06
    imum
    0.06
    aney
    0.06
    ipy
    0.06
     retrofit
    0.06
    Activations Density 0.013%

    No Known Activations