Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 91887
    Prev
    Next
    INDEX
    Explanations

    legal documents

    np_max-act · gemini-2.0-flash

    The neuron activates on the month name in document dates (e.g. “April,” “February”) indicating it’s detecting the month component of a date.

    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
     weekdays
    -0.07
     ['/
    -0.07
    _return
    -0.06
     ethical
    -0.06
     фил
    -0.06
     pages
    -0.06
     dương
    -0.06
     İç
    -0.06
    제
    -0.06
    _equals
    -0.06
    POSITIVE LOGITS
     erotisch
    0.07
    _clk
    0.07
     завер
    0.06
     Viking
    0.06
    OperationContract
    0.06
     FreeBSD
    0.06
     steadfast
    0.06
    erli
    0.06
     cheering
    0.06
    merchant
    0.06
    Activations Density 0.003%

    No Known Activations