© 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. 44819
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    The neuron activates on words and short phrases that express possibility or ability (modal/hedging language such as “may,” “can,” “possible to,” etc.).

    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
    _channel
    -0.06
    ibar
    -0.06
     Riley
    -0.06
    Wie
    -0.06
     Ali
    -0.06
     siempre
    -0.06
     otherwise
    -0.06
     Fulton
    -0.06
     zeros
    -0.06
    .gridView
    -0.05
    POSITIVE LOGITS
    ！↵↵
    0.07
    σταση
    0.06
    нос
    0.06
    -readable
    0.06
    _MATH
    0.06
    jquery
    0.06
     endforeach
    0.06
    .builder
    0.06
    Cfg
    0.06
    !");↵
    0.06
    Activations Density 0.149%

    No Known Activations