© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 91141
    Prev
    Next
    INDEX
    Explanations

    Connecticut

    np_max-act · gemini-2.0-flash

    The neuron strongly responds to occurrences of the jurisdiction name “Connecticut” (including its split-up sub-tokens and the abbreviation “CT”).

    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
     Rod
    -0.07
     conspicuous
    -0.07
     VGA
    -0.07
     ner
    -0.07
    Lady
    -0.06
    getLocation
    -0.06
    มา
    -0.06
     Lola
    -0.06
    !»
    -0.06
     Percy
    -0.06
    POSITIVE LOGITS
     Connecticut
    0.10
     Yale
    0.08
     Vermont
    0.07
     CT
    0.07
    CompatActivity
    0.07
     Domain
    0.07
     ql
    0.07
    activ
    0.07
    .breakpoints
    0.07
    -transitional
    0.07
    Activations Density 0.008%

    No Known Activations