© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 44498
    Prev
    Next
    INDEX
    Explanations

    possessives/contractions

    np_max-act · gemini-2.0-flash

    This neuron selectively responds to capitalized tokens in named entities (e.g. proper nouns like “Hotel,” “Central,” and possessive apostrophes in names like Paris’ or Manhattan’s).

    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
    igma
    -0.07
     humidity
    -0.07
    Meeting
    -0.07
     YEAR
    -0.07
     år
    -0.07
     admit
    -0.07
     činnosti
    -0.06
     Day
    -0.06
     licence
    -0.06
    ابی
    -0.06
    POSITIVE LOGITS
    BeforeEach
    0.07
    (strict
    0.07
    \Carbon
    0.06
     zřejmě
    0.06
     Separator
    0.06
     NYPD
    0.06
    stdbool
    0.06
     огром
    0.06
    nn
    0.06
     ESV
    0.06
    Activations Density 0.018%

    No Known Activations