© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 68840
    Prev
    Next
    INDEX
    Explanations

    9

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on the byte‐escape fragments (e.g. “xe2”, “x80”, “x9c”, “x9d”) that represent UTF-8–encoded non-ASCII punctuation in the text.

    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
    ernels
    -0.08
    atform
    -0.07
    鉄
    -0.06
    Ast
    -0.06
    (conv
    -0.06
    <Func
    -0.06
     řed
    -0.06
    omy
    -0.06
    artists
    -0.06
    .Release
    -0.06
    POSITIVE LOGITS
    COLUMN
    0.07
     procedural
    0.06
     одне
    0.06
     Pl
    0.06
     الات
    0.06
     Marino
    0.06
    ["
    0.06
    readystatechange
    0.06
     UserControl
    0.06
     spou
    0.06
    Activations Density 0.002%

    No Known Activations