© 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. 128884
    Prev
    Next
    INDEX
    Explanations

    Imagine

    np_max-act · gemini-2.0-flash

    The neuron detects expressions of personal worry or concern (e.g. “I worry about…”).

    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
    .Parser
    -0.07
     Warfare
    -0.06
    adena
    -0.06
     ($(
    -0.06
    otten
    -0.06
    ￣￣
    -0.06
    Bi
    -0.06
    本当
    -0.06
     Playlist
    -0.06
    Refresh
    -0.06
    POSITIVE LOGITS
    _HIGH
    0.07
     preval
    0.06
    린
    0.06
     Dutch
    0.06
    opper
    0.06
     DWC
    0.06
     وب
    0.06
     khó
    0.05
    (cmd
    0.05
    ++;
    0.05
    Activations Density 0.027%

    No Known Activations