© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 68689
    Prev
    Next
    INDEX
    Explanations

    filler words, punctuation

    np_max-act · gemini-2.0-flash

    This neuron detects self-deprecating or vulnerability-expressing words (negative self-assessment or emotional insecurity terms).

    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
     Isaac
    -0.06
    mouseleave
    -0.06
    union
    -0.06
     murm
    -0.06
    标题
    -0.06
    енти
    -0.06
    K
    -0.06
    Also
    -0.06
     beneath
    -0.06
    дон
    -0.05
    POSITIVE LOGITS
     Playback
    0.07
    Condition
    0.07
    .Dialog
    0.06
    شي
    0.06
    .Usuario
    0.06
    .semantic
    0.06
     "{
    0.06
     Riding
    0.06
    EObject
    0.06
    важа
    0.06
    Activations Density 0.032%

    No Known Activations