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. 55178
    Prev
    Next
    INDEX
    Explanations

    re

    np_max-act · gemini-2.0-flash

    The main thing this neuron does is detect multi-part technical terms—especially hyphenated compounds and acronyms—that are typical of structured or formal writing.

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

    medical terms related to physical health and strength assessments.

    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
    income
    -0.07
    Configs
    -0.06
    _metrics
    -0.06
    Ž
    -0.06
    aters
    -0.06
    .setLevel
    -0.06
    death
    -0.06
     stickers
    -0.06
    ół
    -0.06
    Camp
    -0.05
    POSITIVE LOGITS
    ΑΜ
    0.07
    _relations
    0.07
    작
    0.07
    lod
    0.07
     spep
    0.07
     GER
    0.06
    弄
    0.06
     여러
    0.06
    LIK
    0.06
     Sink
    0.06
    Activations Density 0.250%

    No Known Activations