© 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 (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 30470
    Prev
    Next
    INDEX
    Explanations

    encyclopedia entries

    np_max-act · gemini-2.0-flash

    This neuron fires on key topical-domain nouns or noun phrases (e.g. “solar power,” “Roman Catholicism,” “Eastern Orthodoxy,” “cigarette,” “Spanish language”) that label the subject or category of an article.

    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
    Ин
    -0.06
    куп
    -0.06
     bà
    -0.06
     pending
    -0.06
     degree
    -0.06
     sharpen
    -0.06
    이에
    -0.06
     dönem
    -0.06
     즉
    -0.06
     accessory
    -0.06
    POSITIVE LOGITS
    weetalert
    0.07
    수
    0.07
    890
    0.06
     afar
    0.06
    fdf
    0.06
    					  
    0.06
     ht
    0.06
     ellipse
    0.06
     endiş
    0.06
     conect
    0.06
    Activations Density 0.016%

    No Known Activations