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. 40602
    Prev
    Next
    INDEX
    Explanations

    The

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on the capitalized definite article “The,” especially when it appears at the start of a titled section.

    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
     Neuroscience
    -0.07
     Briggs
    -0.07
    roller
    -0.07
     PDT
    -0.06
    oleč
    -0.06
     IPA
    -0.06
    autorelease
    -0.06
     Pavilion
    -0.06
    ILITY
    -0.06
    (Mouse
    -0.06
    POSITIVE LOGITS
     benöt
    0.07
    <!--<
    0.06
    扩
    0.06
     Join
    0.06
    śli
    0.06
     چت
    0.06
     luck
    0.06
    hg
    0.06
    shall
    0.06
    _vm
    0.06
    Activations Density 0.001%

    No Known Activations