© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 128842
    Prev
    Next
    INDEX
    Explanations

    Encoded data

    np_max-act · gemini-2.0-flash

    This neuron activates on substrings characteristic of Base64‐encoded data (long runs of A–Z, a–z, 0–9, “+”/“/”, or “=”).

    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
    "h
    -0.07
    وده
    -0.07
    َي
    -0.06
    igit
    -0.06
     Video
    -0.06
     reef
    -0.06
    卢
    -0.06
    gi
    -0.06
     metric
    -0.06
    esity
    -0.06
    POSITIVE LOGITS
     frontline
    0.07
    _fixed
    0.06
    !");
    0.06
     listOf
    0.06
    征
    0.06
     VStack
    0.06
     TKey
    0.06
     rewards
    0.06
    subscriptions
    0.06
    .herokuapp
    0.06
    Activations Density 0.003%

    No Known Activations