© 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. 12732
    Prev
    Next
    INDEX
    Explanations

    door

    np_max-act · gemini-2.0-flash

    The neuron fires on simple spatial/action verbs that describe moving through or interacting with the environment (e.g. “walked,” “shut,” “onto,” “out”).

    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
    ?“
    -0.07
    ))"↵
    -0.07
    292
    -0.07
     berry
    -0.06
    Username
    -0.06
    pNet
    -0.06
    182
    -0.06
    ograph
    -0.06
    ]<<"
    -0.06
     blas
    -0.06
    POSITIVE LOGITS
     habit
    0.06
     states
    0.06
    owej
    0.06
     surrounding
    0.06
    notif
    0.06
    vasive
    0.06
    .weather
    0.06
    �
    0.06
    datetime
    0.06
    ,tr
    0.05
    Activations Density 0.058%

    No Known Activations