© 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. 13422
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron responds to tokens in metadata‐style information—especially numbers, dates, and common connectors (e.g. “and,” “in,” “.”) associated with release details.

    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
    xz
    -0.07
     PED
    -0.06
     Chad
    -0.06
     Pig
    -0.06
     york
    -0.06
    Dependency
    -0.06
    stances
    -0.06
     Ruby
    -0.06
     claim
    -0.06
     usa
    -0.06
    POSITIVE LOGITS
    >'.
    0.08
    _CHARSET
    0.07
     dál
    0.07
    .executeQuery
    0.07
    <Scalars
    0.06
     vua
    0.06
    rowse
    0.06
    .addHandler
    0.06
     '">'
    0.06
     тисяч
    0.06
    Activations Density 0.032%

    No Known Activations