© 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. 92631
    Prev
    Next
    INDEX
    Explanations

    are

    np_max-act · gemini-2.0-flash

    This neuron detects the boilerplate data‐availability statement (e.g. the “All relevant data are …” sentence at the start of a manuscript).

    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
    hci
    -0.07
    date
    -0.07
    ')){↵
    -0.06
    lea
    -0.06
     header
    -0.06
    required
    -0.06
    _subject
    -0.06
    уст
    -0.06
     bulletin
    -0.06
    785
    -0.06
    POSITIVE LOGITS
     trabajar
    0.06
     elabor
    0.06
     grandes
    0.06
     TValue
    0.06
     infr
    0.06
    γκό
    0.06
     φω
    0.06
     delve
    0.06
     irrespective
    0.06
     trotz
    0.06
    Activations Density 0.000%

    No Known Activations