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. 85931
    Prev
    Next
    INDEX
    Explanations

    a

    np_max-act · gemini-2.0-flash

    The neuron detects figure subpanel labels (e.g. the “A,” “B,” etc. in “Fig. 1A,” “Fig. 1B,” etc.).

    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
    _FALL
    -0.08
    .Cons
    -0.07
    Seen
    -0.07
    =datetime
    -0.07
     pisc
    -0.07
    Values
    -0.07
    .owl
    -0.06
    .Age
    -0.06
    esterday
    -0.06
    _LT
    -0.06
    POSITIVE LOGITS
    числ
    0.07
     intermedi
    0.07
     الز
    0.06
    containers
    0.06
     استرات
    0.06
     Supplementary
    0.06
     supplementary
    0.06
     ozone
    0.06
    ulla
    0.06
     UIColor
    0.06
    Activations Density 0.002%

    No Known Activations