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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 8721
    Prev
    Next
    INDEX
    Explanations

    period

    np_max-act · gemini-2.0-flash

    The neuron is activated by floating‐point numeric literals (decimal numbers) in the text.

    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
    trained
    -0.07
     fing
    -0.07
    faculty
    -0.07
     普通
    -0.07
     Соб
    -0.06
     spoof
    -0.06
     bunny
    -0.06
     scheduler
    -0.06
     พร
    -0.06
    isplay
    -0.06
    POSITIVE LOGITS
     maduras
    0.06
    (dot
    0.06
     stemming
    0.06
    관련
    0.06
    heten
    0.06
     resin
    0.06
     stalo
    0.06
     etkil
    0.06
     determ
    0.06
     necessário
    0.06
    Activations Density 0.029%

    No Known Activations