© 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. 110192
    Prev
    Next
    INDEX
    Explanations

    something being finished

    np_max-act · gemini-2.0-flash

    The neuron fires on first-person expressions of eagerness for something (a day, week, etc.) to end—for example “I just want this week to be over.”

    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
    istogram
    -0.07
     taxis
    -0.07
    dal
    -0.07
    аря
    -0.06
     arrangement
    -0.06
     hoog
    -0.06
    _run
    -0.06
     flour
    -0.06
    Protect
    -0.06
     boosts
    -0.06
    POSITIVE LOGITS
     Bihar
    0.06
    >>(↵
    0.06
     방법
    0.06
    agem
    0.06
    ível
    0.06
     vb
    0.06
    ’l
    0.06
     distancia
    0.06
    OTT
    0.06
     lingering
    0.06
    Activations Density 0.023%

    No Known Activations