© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 92014
    Prev
    Next
    INDEX
    Explanations

    ly

    np_max-act · gemini-2.0-flash

    The neuron activates on descriptive adjectives and adverbs that qualify quality or degree (e.g. “relevant,” “precise,” “stable,” “fluidly”).

    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
     hard
    -0.08
     big
    -0.07
     very
    -0.07
     little
    -0.06
     only
    -0.06
    formatter
    -0.06
     other
    -0.06
    πλα
    -0.06
    .phoneNumber
    -0.06
     вер
    -0.06
    POSITIVE LOGITS
     출
    0.07
     "#{
    0.06
     Olympic
    0.06
     padr
    0.06
    律
    0.06
    $field
    0.06
    onal
    0.06
    Rol
    0.06
    naments
    0.06
    temps
    0.06
    Activations Density 0.459%

    No Known Activations