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. 84764
    Prev
    Next
    INDEX
    Explanations

    instructions

    np_max-act · gemini-2.0-flash

    This neuron responds to terms describing the selection of the first or sole match (e.g. “first,” “only,” “matching,” “returned”) in function behavior or documentation.

    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
    utter
    -0.07
    uslim
    -0.07
    namespace
    -0.07
     porter
    -0.06
     lisans
    -0.06
    `
    ↵
    -0.06
     نظامی
    -0.06
     constructors
    -0.06
     shards
    -0.06
    ByVersion
    -0.06
    POSITIVE LOGITS
    Tpl
    0.07
     zach
    0.07
    .os
    0.07
    .are
    0.06
     odor
    0.06
    —but
    0.06
    -facing
    0.06
     Hyp
    0.06
    ^(
    0.06
     divul
    0.06
    Activations Density 0.037%

    No Known Activations