© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 38280
    Prev
    Next
    INDEX
    Explanations

    Conversational text

    np_max-act · gemini-2.0-flash

    instances of casual conversation and humor in communication.

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

    The neuron reliably lights up on pronouns and other small deictic words (e.g. I, me, you, she, that, lo) indicating personal or discourse reference.

    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
    Pool
    -0.07
    	X
    -0.07
    ิโล
    -0.06
     Imported
    -0.06
     руку
    -0.06
     kullanarak
    -0.06
    -core
    -0.06
     GetHashCode
    -0.06
    .AP
    -0.06
    	grid
    -0.06
    POSITIVE LOGITS
    都市
    0.07
    .FullName
    0.06
     olig
    0.06
    Sites
    0.06
    orelease
    0.06
    .Filter
    0.06
    ==
    0.06
    _BOTTOM
    0.06
     LESS
    0.06
     wol
    0.06
    Activations Density 0.076%

    No Known Activations