© 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. 68776
    Prev
    Next
    INDEX
    Explanations

    FIG

    np_max-act · gemini-2.0-flash

    The neuron activates on numerical labels and figure/section references (e.g. “1.”, “FIG 3.x”, “Referring to FIG…”), i.e. it detects numbered headings and figure citations.

    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
    ModelState
    -0.07
     Older
    -0.06
    ��제
    -0.06
    _term
    -0.06
    冠
    -0.06
    eker
    -0.06
     Short
    -0.06
    スカ
    -0.06
     L
    -0.06
     userModel
    -0.06
    POSITIVE LOGITS
    <decltype
    0.07
     Nokia
    0.07
     Leonardo
    0.06
     hãng
    0.06
    /octet
    0.06
     akşam
    0.06
    *g
    0.06
    (pb
    0.06
    -Nov
    0.06
    Greetings
    0.06
    Activations Density 0.002%

    No Known Activations