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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 327
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash

    This neuron detects temporal references—mentions of dates, years, centuries/decades, and ordinal time periods (e.g. “first,” “20th century,” “World War I”).

    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
     lobbyist
    -0.07
    .ping
    -0.07
    DOMContentLoaded
    -0.07
     subsidy
    -0.07
     lawyer
    -0.07
    يرة
    -0.07
    irthday
    -0.06
    .handlers
    -0.06
    Ky
    -0.06
    opal
    -0.06
    POSITIVE LOGITS
     tweeting
    0.06
    так
    0.06
    ******↵↵
    0.06
    ITIONAL
    0.06
     homeowner
    0.06
     laughed
    0.06
     incapac
    0.06
     dej
    0.06
     modele
    0.06
     ContentValues
    0.06
    Activations Density 0.012%

    No Known Activations