© 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. 8927
    Prev
    Next
    INDEX
    Explanations

    a

    np_max-act · gemini-2.0-flash

    This neuron activates on the user’s self-descriptions of ongoing work—specifically the “I am working on a …” or “on a …” phrasing that kicks off a project statement.

    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
     sandwich
    -0.07
    .throw
    -0.06
    Worker
    -0.06
     photographer
    -0.06
    erson
    -0.06
    Binder
    -0.06
    Inflater
    -0.06
     Hindered
    -0.06
     Goth
    -0.06
    .Pro
    -0.06
    POSITIVE LOGITS
     ма
    0.07
     silence
    0.06
    曜
    0.06
     tá
    0.06
    	ans
    0.06
    commission
    0.06
     twink
    0.06
     Fe
    0.06
    τες
    0.06
     plea
    0.06
    Activations Density 0.162%

    No Known Activations