© 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. 12939
    Prev
    Next
    INDEX
    Explanations

    math

    np_max-act · gemini-2.0-flash

    The neuron flags occurrences of “sum of” in arithmetic questions—that is, it detects when the text is asking to compute a sum.

    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
    _rows
    -0.08
    West
    -0.06
     inherits
    -0.06
     People
    -0.06
     Ваш
    -0.06
     args
    -0.06
    -eyed
    -0.06
    aisy
    -0.06
    people
    -0.06
     WHITE
    -0.06
    POSITIVE LOGITS
     водой
    0.07
    .getInfo
    0.06
    /mol
    0.06
     remainder
    0.06
     podařilo
    0.06
    /tcp
    0.06
     stating
    0.06
    既
    0.06
    _rho
    0.06
     شر
    0.06
    Activations Density 0.012%

    No Known Activations