© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 109894
    Prev
    Next
    INDEX
    Explanations

    hold

    np_max-act · gemini-2.0-flash

    This neuron detects second-person reader-addressing phrases describing the reader’s actions (e.g. “you move your hand,” “you take a closer look,” “hold it just right”).

    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
    Edges
    -0.07
     Moto
    -0.07
     Builders
    -0.07
    Topology
    -0.07
    Für
    -0.06
    Lisa
    -0.06
     alın
    -0.06
    Krist
    -0.06
     analyzer
    -0.06
    Personally
    -0.06
    POSITIVE LOGITS
     %
    0.07
    	old
    0.07
     Personen
    0.06
     国家
    0.06
    емого
    0.06
    opi
    0.06
    [iVar
    0.06
     centr
    0.06
    eres
    0.06
     substitute
    0.06
    Activations Density 0.049%

    No Known Activations