© 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. 12683
    Prev
    Next
    INDEX
    Explanations

    200

    np_max-act · gemini-2.0-flash

    This neuron activates on the “200” token in date stamps, i.e. the year prefix of four-digit dates like “2007,” “2008,” “2009,” etc.

    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
     норматив
    -0.07
    ane
    -0.06
    IColor
    -0.06
     největší
    -0.06
    IQ
    -0.06
    Finite
    -0.06
    預
    -0.06
    スティ
    -0.06
     getApp
    -0.06
    infile
    -0.06
    POSITIVE LOGITS
     THEM
    0.07
     등록
    0.07
     lemon
    0.07
    說
    0.07
    (mark
    0.07
     overwhelmingly
    0.07
     premium
    0.06
    (asset
    0.06
    cart
    0.06
     Vish
    0.06
    Activations Density 0.001%

    No Known Activations