© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 89800
    Prev
    Next
    INDEX
    Explanations

    said

    np_max-act · gemini-2.0-flash

    The neuron fires on words that signal reported speech or statements—i.e. verbs like “said,” “stated,” “informed,” “asked,” 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
     Portsmouth
    -0.07
     Phones
    -0.07
     Ukrainian
    -0.06
    이다
    -0.06
    �
    -0.06
    가요
    -0.06
     Kanye
    -0.06
    ilit
    -0.06
    homes
    -0.06
     applied
    -0.06
    POSITIVE LOGITS
    _group
    0.07
    ця
    0.07
    sym
    0.06
    .world
    0.06
    meler
    0.06
     relent
    0.06
    plete
    0.06
    Voice
    0.06
    rection
    0.06
     PHYS
    0.06
    Activations Density 0.047%

    No Known Activations