© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 88905
    Prev
    Next
    INDEX
    Explanations

    knowing

    np_max-act · gemini-2.0-flash

    subject matter related to medical diagnostic studies and their outcomes.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron detects mentions of blinded assessments—i.e. words signaling that evaluators or observers are “not aware,” “without knowing,” “unaware,” or otherwise masked from key information.

    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
     RAD
    -0.07
     own
    -0.07
     WAL
    -0.07
     getSupportFragmentManager
    -0.06
    قیق
    -0.06
     Powell
    -0.06
    elix
    -0.06
    знач
    -0.06
     tantra
    -0.06
    olest
    -0.06
    POSITIVE LOGITS
    ��
    0.06
     Pend
    0.06
    .isNull
    0.06
     Düz
    0.06
    atis
    0.06
     potentially
    0.06
     POD
    0.06
    ुए
    0.06
     shiny
    0.06
    ตะว
    0.06
    Activations Density 0.005%

    No Known Activations