© 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. 7897
    Prev
    Next
    INDEX
    Explanations

    technical descriptions

    np_max-act · gemini-2.0-flash

    This neuron detects the special role‐label tokens marking the assistant’s header (e.g. the `<|start_header_id|>` … `assistant` tag).

    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
     disciples
    -0.07
     lightweight
    -0.07
    inar
    -0.06
    ;'↵
    -0.06
    스트
    -0.06
    _ITEM
    -0.06
     breathed
    -0.06
    INTR
    -0.06
    ENSION
    -0.06
    bír
    -0.06
    POSITIVE LOGITS
     tum
    0.06
    Rectangle
    0.06
     शहर
    0.06
     createDate
    0.06
    SupportFragmentManager
    0.06
     Examples
    0.06
     Kam
    0.06
    .cum
    0.06
     lối
    0.06
     caffe
    0.06
    Activations Density 0.043%

    No Known Activations