© 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. 88941
    Prev
    Next
    INDEX
    Explanations

    classifications

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens (especially classification digits like the “2” in “Category 2 carcinogen”).

    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
    ЕС
    -0.07
     Вас
    -0.07
    _des
    -0.06
    _connection
    -0.06
    xa
    -0.06
    Fig
    -0.06
    統
    -0.06
    dro
    -0.06
    艇
    -0.06
    erte
    -0.06
    POSITIVE LOGITS
    .Pool
    0.06
    ASSES
    0.06
    .createStatement
    0.06
     gece
    0.06
    -now
    0.06
    <body
    0.06
     NORTH
    0.06
     beberapa
    0.06
    (月
    0.06
    _TIMEOUT
    0.06
    Activations Density 0.007%

    No Known Activations