© 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. 728
    Prev
    Next
    INDEX
    Explanations

    descriptions

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of “used” (and related passive‐voice constructions like “is used,” “are used”) that describe how something is applied.

    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
    .Designer
    -0.07
     precursor
    -0.06
     SSP
    -0.06
     Trafford
    -0.06
    _packages
    -0.06
     серд
    -0.06
    MX
    -0.06
    ителей
    -0.06
    X
    -0.06
    ocuk
    -0.06
    POSITIVE LOGITS
    喊
    0.07
    propTypes
    0.07
    _STMT
    0.06
     lighter
    0.06
    kon
    0.06
     prune
    0.06
    TextView
    0.06
     ένας
    0.06
    енты
    0.06
    까
    0.06
    Activations Density 0.041%

    No Known Activations