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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 44034
    Prev
    Next
    INDEX
    Explanations

    "'s" and "is"

    np_max-act · gemini-2.0-flash

    This neuron selectively detects the contraction “‘s” (as in “there’s,” “that’s,” 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
     cou
    -0.07
    Compra
    -0.06
     Santa
    -0.06
     Hãy
    -0.06
     better
    -0.06
     ImageButton
    -0.06
     Fonts
    -0.06
     Sachs
    -0.06
    .FindAsync
    -0.06
     Sick
    -0.06
    POSITIVE LOGITS
    posta
    0.07
     dignity
    0.06
    possible
    0.06
    crement
    0.06
    uffle
    0.06
    :C
    0.06
    Del
    0.06
     honorable
    0.06
    !");
    ↵
    0.06
    Mi
    0.06
    Activations Density 0.021%

    No Known Activations