© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 128649
    Prev
    Next
    INDEX
    Explanations

    Unstructured data

    np_max-act · gemini-2.0-flash

    This neuron spotlights the formal headings and citations in legal texts—section numbers, titles (e.g. “SHORT TITLE,” “SECTION”), Act names, dates, and related statutory reference tokens.

    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
     GroupLayout
    -0.07
    ])↵↵
    -0.06
     Geç
    -0.06
    `.↵↵
    -0.06
    ыш
    -0.06
    ssql
    -0.06
    ϊκ
    -0.06
     Nhưng
    -0.05
     přep
    -0.05
    ũng
    -0.05
    POSITIVE LOGITS
    review
    0.07
     StartTime
    0.06
     Underground
    0.06
     DXGI
    0.06
     PRES
    0.06
     клет
    0.06
     Oasis
    0.06
    _decl
    0.06
     acompan
    0.06
     composite
    0.06
    Activations Density 0.022%

    No Known Activations