Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 92785
    Prev
    Next
    INDEX
    Explanations

    types of government

    np_max-act · gemini-2.0-flash

    The neuron activates on words naming forms or elements of governance—e.g. “political,” “regal,” “civil,” “domestic,” “family,” “government,” etc.

    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
     grave
    -0.07
    üğ
    -0.07
    _rad
    -0.07
    으니
    -0.06
     jiný
    -0.06
    adě
    -0.06
     crystals
    -0.06
     ceiling
    -0.06
    nonce
    -0.06
     mis
    -0.06
    POSITIVE LOGITS
    技术
    0.07
    *↵↵
    0.06
     Classics
    0.06
    _DDR
    0.06
     entreprise
    0.06
     Οικο
    0.06
    ANEL
    0.06
    _EVAL
    0.06
    ORIGINAL
    0.06
    dbcTemplate
    0.06
    Activations Density 0.000%

    No Known Activations