© 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. 125038
    Prev
    Next
    INDEX
    Explanations

    learn

    np_max-act · gemini-2.0-flash

    The neuron identifies phrases that refer to applying or having acquired knowledge and skills (e.g. mentions of “I,” “learned,” “skills,” “knowledge”).

    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
    Screens
    -0.07
    ホテル
    -0.07
    SELECT
    -0.07
     dusty
    -0.07
     Charm
    -0.06
     orbital
    -0.06
     Moose
    -0.06
     filme
    -0.06
     SCREEN
    -0.06
    ########################################
    -0.06
    POSITIVE LOGITS
     conserv
    0.07
     "\""
    0.07
     tuổi
    0.07
     addressed
    0.06
     hesitant
    0.06
    OrDefault
    0.06
    .Tree
    0.06
     thật
    0.06
     Кол
    0.06
     istediğiniz
    0.06
    Activations Density 0.034%

    No Known Activations