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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 7861
    Prev
    Next
    INDEX
    Explanations

    TAG

    np_max-act · gemini-2.0-flash

    The neuron is specialized to pick up on “TAG” labels in metadata headers (e.g. the %TAG and tag fields in Unity/YAML files).

    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
    鲜
    -0.06
     Hwy
    -0.06
     Winn
    -0.06
    κος
    -0.06
     σκο
    -0.06
     realpath
    -0.06
    _primitive
    -0.06
    비스
    -0.06
     giov
    -0.06
    ecimal
    -0.06
    POSITIVE LOGITS
    .ct
    0.07
    -valid
    0.07
    IF
    0.07
    TS
    0.06
     delegate
    0.06
    áky
    0.06
     tag
    0.06
    부분
    0.06
     CODE
    0.06
    [token
    0.06
    Activations Density 0.000%

    No Known Activations