© 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. 30519
    Prev
    Next
    INDEX
    Explanations

    headlines and announcements

    np_max-act · gemini-2.0-flash

    This neuron highlights section titles or headings—tokens that belong to article/job‐posting/announcement headers and other metadata presented as capitalized labels (e.g. news section names, job titles, dates, or report titles).

    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
     depletion
    -0.06
     Perry
    -0.06
     typed
    -0.06
    (ds
    -0.06
     Brake
    -0.06
    -cn
    -0.06
    clients
    -0.06
    Creat
    -0.06
     Kb
    -0.06
    sha
    -0.06
    POSITIVE LOGITS
     νεφοκάλυψης
    0.07
     üzerinden
    0.07
     Barber
    0.07
     Velvet
    0.07
     قطع
    0.06
     سفر
    0.06
    .…
    0.06
    	HANDLE
    0.06
    /net
    0.06
    Dice
    0.06
    Activations Density 0.087%

    No Known Activations