© 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. 90827
    Prev
    Next
    INDEX
    Explanations

    ificance

    np_max-act · gemini-2.0-flash

    The neuron fires on JSON field names in the assistant’s structured output (i.e. tokens like “professional,” “goal,” “special,” “skills,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    professionals related to historical and scientific contexts, specifically those skilled in mathematics, physics, or history.

    oai_token-act-pair · gpt-4o-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
    flu
    -0.07
     ],
↵
    -0.06
     возраст
    -0.06
    ازات
    -0.06
    Power
    -0.06
    печ
    -0.06
     střed
    -0.06
    foundland
    -0.06
    /<?
    -0.06
    _TEXT
    -0.06
    POSITIVE LOGITS
     Gri
    0.07
     __('
    0.06
     extra
    0.06
     setw
    0.06
    公式
    0.06
    (repo
    0.06
     Bris
    0.06
     Prec
    0.06
     Від
    0.06
    	Send
    0.06
    Activations Density 0.015%

    No Known Activations