© 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. 6815
    Prev
    Next
    INDEX
    Explanations

    }

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numeric and formatting tokens—section numbers, floating‐point measurements, and table/grid delimiters—i.e. it detects structured numerical data and headings.

    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
    \Carbon
    -0.07
    Unique
    -0.06
    .LogWarning
    -0.06
    べて
    -0.06
    ubic
    -0.06
    .GenerationType
    -0.06
     Zion
    -0.06
    \Middleware
    -0.06
    grammar
    -0.06
    .vector
    -0.06
    POSITIVE LOGITS
    ANCED
    0.07
     CLLocationCoordinate
    0.06
     사이
    0.06
     интерес
    0.06
    死
    0.06
     جه
    0.06
    ATING
    0.06
     \\
    0.06
    (draw
    0.06
    пп
    0.06
    Activations Density 0.003%

    No Known Activations