© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 73822
    Prev
    Next
    INDEX
    Explanations

    technical processes

    np_max-act · gemini-2.0-flash

    This neuron responds to past-participle verb forms (e.g. “made,” “set,” “covered,” “connected”) describing processes or states in technical descriptions.

    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
     Doch
    -0.07
     Arsenal
    -0.06
     EntityManager
    -0.06
    (predictions
    -0.06
    _compiler
    -0.06
     Pete
    -0.06
     Deleting
    -0.06
    _servers
    -0.06
     schw
    -0.06
    beh
    -0.06
    POSITIVE LOGITS
    acion
    0.07
    ation
    0.07
    ación
    0.07
    .ends
    0.07
     відпов
    0.07
    Statics
    0.07
    ्दर
    0.06
     تصویر
    0.06
    ACION
    0.06
     выступ
    0.06
    Activations Density 0.071%

    No Known Activations