© 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. 43352
    Prev
    Next
    INDEX
    Explanations

    technical writing

    np_max-act · gemini-2.0-flash

    This neuron responds to uncommon, domain‐specific multi‐token terms (e.g. “sleep apnea,” “whiplash,” “ASTM headgear,” “selectively actuated,” etc.).

    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
     producto
    -0.06
     мен
    -0.06
     affects
    -0.06
     각
    -0.06
     cal
    -0.06
     Beats
    -0.06
     tack
    -0.06
     parties
    -0.06
    .Click
    -0.06
    ucus
    -0.06
    POSITIVE LOGITS
     JMP
    0.07
    sendMessage
    0.07
    驚
    0.07
    imest
    0.06
    _translation
    0.06
    exception
    0.06
    [t
    0.06
     Occupational
    0.06
    鏡
    0.06
    šť
    0.06
    Activations Density 0.349%

    No Known Activations