© 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. 110171
    Prev
    Next
    INDEX
    Explanations

    post

    np_max-act · gemini-2.0-flash

    The neuron detects bits of Hacker News submission metadata—things like “submitted,” vote/point counts, and time‐ago timestamps.

    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
    キ
    -0.06
     EA
    -0.06
    _IP
    -0.06
    iye
    -0.06
    eden
    -0.06
    Name
    -0.06
    戒
    -0.06
    ques
    -0.06
     Encyclopedia
    -0.06
     OBS
    -0.06
    POSITIVE LOGITS
    sgiving
    0.07
     Hindus
    0.06
    	background
    0.06
     Gothic
    0.06
     fiscal
    0.06
     Tradable
    0.06
    ummings
    0.06
    ]*
    0.06
    PDOException
    0.06
    placer
    0.06
    Activations Density 0.009%

    No Known Activations