Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 130185
    Prev
    Next
    INDEX
    Explanations

    number

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens or sequences of digits (i.e. it flags occurrences of numbers in the text).

    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
     RectTransform
    -0.06
     jan
    -0.06
     ------
    -0.06
    bia
    -0.06
    (Screen
    -0.06
     \
    -0.06
     시간
    -0.06
    _serv
    -0.06
    ょう
    -0.05
    ‌کنندگان
    -0.05
    POSITIVE LOGITS
     disturbances
    0.07
     kart
    0.07
     полез
    0.07
     disturbance
    0.06
    icit
    0.06
    -inspired
    0.06
     skup
    0.06
     resent
    0.06
    adresse
    0.06
     werk
    0.06
    Activations Density 0.015%

    No Known Activations