© 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. 126359
    Prev
    Next
    INDEX
    Explanations

    disagreements

    np_max-act · gemini-2.0-flash

    The neuron specifically fires on the word “odds” (as in “at odds”), flagging instances of that contrastive expression.

    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
    елов
    -0.07
     Тим
    -0.06
     when
    -0.06
     elabor
    -0.06
     debugging
    -0.06
    (Void
    -0.06
     Alma
    -0.06
     contemplated
    -0.06
    必要
    -0.06
     Allowed
    -0.06
    POSITIVE LOGITS
    Genres
    0.06
    (rc
    0.06
    "testing
    0.06
    issen
    0.06
    siz
    0.06
    àu
    0.06
    кот
    0.06
     bestellen
    0.06
     fq
    0.06
     Quant
    0.06
    Activations Density 0.048%

    No Known Activations