© 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. 91958
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron fires on forum‐style quote blocks or anything marked out as a quotation (e.g. lines introduced by “>”, “Re:”, opening and closing quotes, or the dashed/underlined separators that denote quoted 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
    .do
    -0.07
    าน
    -0.06
    今
    -0.06
     �
    -0.06
    posting
    -0.06
     ff
    -0.06
    haled
    -0.06
    dist
    -0.06
    /no
    -0.06
     nest
    -0.06
    POSITIVE LOGITS
    Sept
    0.07
     [:
    0.06
    \"");↵
    0.06
     ejaculation
    0.06
    "http
    0.06
    ":"","
    0.06
    0.06
    ']),
    0.06
    ड़
    0.06
    вается
    0.06
    Activations Density 0.012%

    No Known Activations