© 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. 42767
    Prev
    Next
    INDEX
    Explanations

    attorney

    np_max-act · gemini-2.0-flash

    This neuron fires on tokens appearing in the case‐caption or counsel listing (e.g. names and titles of lawyers and attorneys).

    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
     AM
    -0.07
     informací
    -0.07
     Cubs
    -0.07
    _CLOSED
    -0.07
     unanswered
    -0.07
    #c
    -0.07
    .ap
    -0.07
    acs
    -0.06
    ività
    -0.06
     ایرانی
    -0.06
    POSITIVE LOGITS
     slag
    0.06
     Spec
    0.06
    );$
    0.05
    idente
    0.05
    iction
    0.05
    umin
    0.05
    __))↵
    0.05
    }")]↵
    0.05
     obsahuje
    0.05
     reson
    0.05
    Activations Density 0.012%

    No Known Activations