© 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. 12777
    Prev
    Next
    INDEX
    Explanations

    plus

    np_max-act · gemini-2.0-flash

    The neuron is triggered by numeric expressions—especially numbers (including decimals) that quantify experience levels, durations, or other specifications—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
     Polit
    -0.07
    rolling
    -0.07
     desperate
    -0.06
     sci
    -0.06
     emotions
    -0.06
    _DIG
    -0.06
     Streets
    -0.06
    шими
    -0.06
    _packet
    -0.06
     Wig
    -0.06
    POSITIVE LOGITS
    ło
    0.07
    び
    0.06
    mesi
    0.06
     зали
    0.06
     cung
    0.06
     připom
    0.06
    .Magenta
    0.06
     Synd
    0.06
    ��
    0.06
    .ResponseWriter
    0.06
    Activations Density 0.009%

    No Known Activations