© 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. 126612
    Prev
    Next
    INDEX
    Explanations

    Additional

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of the word “Additional,” as used in section or heading labels (e.g. “Additional Information,” “Additional capabilities,” etc.).

    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
     sort
    -0.09
     runs
    -0.08
     run
    -0.08
    (run
    -0.08
     Lets
    -0.07
     trot
    -0.07
    (Tree
    -0.07
     surely
    -0.07
    ｢
    -0.07
     say
    -0.07
    POSITIVE LOGITS
     additional
    0.14
    additional
    0.09
     Additional
    0.08
    Additional
    0.07
     aesthetic
    0.07
    ์:
    0.07
    extra
    0.07
    ودی
    0.07
     필
    0.07
    DOI
    0.07
    Activations Density 0.016%

    No Known Activations