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. 204
    Prev
    Next
    INDEX
    Explanations

    news articles

    np_max-act · gemini-2.0-flash

    The neuron selectively responds to sequences of uppercase letters (e.g. acronyms, headings, or all-caps words).

    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
    64
    -0.07
    .Box
    -0.07
     lé
    -0.07
    })
    ↵
    ↵
    -0.06
    ?“↵↵
    -0.06
    DTO
    -0.06
     данны
    -0.06
    تن
    -0.06
     суд
    -0.06
    وینت
    -0.06
    POSITIVE LOGITS
     letras
    0.06
    赏
    0.06
    \OptionsResolver
    0.06
    €“
    0.06
    (blocks
    0.06
    indx
    0.06
    'est
    0.06
    ิลป
    0.06
     Biological
    0.06
    .addr
    0.05
    Activations Density 0.013%

    No Known Activations