© 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. 43363
    Prev
    Next
    INDEX
    Explanations

    are

    np_max-act · gemini-2.0-flash

    The neuron fires on common grammatical “glue” words—especially the indefinite article “a,” the infinitival marker “to,” and similarly frequent function words—marking basic English function 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
     Hospitals
    -0.07
    icional
    -0.06
    _CIPHER
    -0.06
     RegExp
    -0.06
     Module
    -0.06
     Prince
    -0.06
     vyvol
    -0.06
    etable
    -0.06
    _abs
    -0.06
     Heal
    -0.05
    POSITIVE LOGITS
    -к
    0.07
     تعریف
    0.07
    _ABI
    0.06
    -member
    0.06
     khắc
    0.06
    lox
    0.06
    
    0.06
     провести
    0.06
    [(
    0.06
     mismatch
    0.06
    Activations Density 0.007%

    No Known Activations