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. 84718
    Prev
    Next
    INDEX
    Explanations

    Legal cases

    np_max-act · gemini-2.0-flash

    The neuron predominantly activates on Westlaw‐style parallel citations (i.e. the “WL” token and its accompanying report numbers).

    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
    mdl
    -0.07
     tore
    -0.07
    :new
    -0.07
     ê
    -0.07
    benchmark
    -0.07
    ,SLOT
    -0.07
     tus
    -0.06
     spoiler
    -0.06
     trips
    -0.06
     stretching
    -0.06
    POSITIVE LOGITS
     FO
    0.07
    lucent
    0.07
    /';↵↵
    0.06
    р
    0.06
    ζα
    0.06
     Sega
    0.06
    others
    0.06
    Declaration
    0.06
    .dst
    0.06
    disciplinary
    0.06
    Activations Density 0.001%

    No Known Activations