© 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. 30116
    Prev
    Next
    INDEX
    Explanations

    articles/prepositions

    np_max-act · gemini-2.0-flash

    This neuron selectively responds to words denoting physical endpoints or spatial relationships—e.g. “ends,” “tips,” “corners,” “adjacent,” and “together.”

    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
    risk
    -0.06
    кости
    -0.06
    rating
    -0.06
     utterly
    -0.06
    Addr
    -0.06
    /Area
    -0.06
     Blank
    -0.06
     Patient
    -0.06
     cloning
    -0.06
    =tmp
    -0.06
    POSITIVE LOGITS
    -Mail
    0.06
    _micro
    0.06
     giz
    0.06
     грав
    0.06
    ()):↵
    0.06
     реак
    0.06
     micron
    0.06
    "
↵
↵
    0.06
    ,[
    0.06
    queueReusable
    0.06
    Activations Density 0.022%

    No Known Activations