© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 92188
    Prev
    Next
    INDEX
    Explanations

    period

    np_max-act · gemini-2.0-flash

    This neuron fires on text that begins new list items or headings—i.e. tokens at the start of numbered or bulleted entries (and the first word following a list marker).

    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
     help
    -0.06
    (e
    -0.06
    *e
    -0.06
    στε
    -0.06
     Desktop
    -0.06
     mouseY
    -0.06
     jobject
    -0.06
    Negative
    -0.06
    inde
    -0.06
     penalty
    -0.06
    POSITIVE LOGITS
    螺
    0.07
    -trade
    0.06
     сили
    0.06
    ести
    0.06
    ﻠ
    0.06
    .pair
    0.06
     reinst
    0.06
    -establish
    0.06
     doprov
    0.06
     neuen
    0.06
    Activations Density 0.125%

    No Known Activations