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. 92950
    Prev
    Next
    INDEX
    Explanations

    Sentence fragments

    np_max-act · gemini-2.0-flash

    The neuron detects Latin‐derived word stems/prefixes (like adv-, trans-, har-, pres-, def-) that signal formal or technical vocabulary in longer 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
     cosmos
    -0.07
     "#"
    -0.07
    ερό
    -0.07
    _AFTER
    -0.06
    ãng
    -0.06
    rawtypes
    -0.06
    ourt
    -0.06
     工
    -0.06
    _LINES
    -0.06
     Kis
    -0.06
    POSITIVE LOGITS
     sami
    0.07
    лод
    0.07
    ีเอ
    0.06
     oyn
    0.06
     voll
    0.06
     uch
    0.06
     아니
    0.06
    Dies
    0.06
     b
    0.06
     치
    0.06
    Activations Density 0.066%

    No Known Activations