© 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. 85783
    Prev
    Next
    INDEX
    Explanations

    spread

    np_max-act · gemini-2.0-flash

    The neuron fires on precise decimal-formatted numeric tokens (e.g. “4.7421875,” “1.892578125”) commonly seen in demographic percentage/statistics tables.

    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
    lining
    -0.07
    นก
    -0.07
     Advisors
    -0.07
    _Buffer
    -0.07
     Masters
    -0.07
     editors
    -0.07
    Apache
    -0.06
    .Match
    -0.06
     purification
    -0.06
    ("%
    -0.06
    POSITIVE LOGITS
     shale
    0.07
    قل
    0.07
     ██
    0.06
    .rdf
    0.06
     canh
    0.06
     prevented
    0.06
    EXPECTED
    0.06
     dovol
    0.06
     collo
    0.06
     Shooter
    0.06
    Activations Density 0.000%

    No Known Activations