© 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. 89127
    Prev
    Next
    INDEX
    Explanations

    said

    np_max-act · gemini-2.0-flash

    statements about political opinions and promises.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    This neuron detects reporting‐speech verbs (e.g. said, insisted, observed, admitted).

    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
    hů
    -0.06
     SIGNAL
    -0.06
     Vision
    -0.06
    ouchers
    -0.06
    га
    -0.06
     intriguing
    -0.06
     Moodle
    -0.06
     производ
    -0.06
    ку
    -0.06
     Cafe
    -0.06
    POSITIVE LOGITS
     pis
    0.07
    된
    0.07
    getElement
    0.06
    希望
    0.06
    (pr
    0.06
     godt
    0.06
     теб
    0.06
     susp
    0.06
     Animated
    0.06
    AccessorType
    0.06
    Activations Density 0.040%

    No Known Activations