Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 86354
    Prev
    Next
    INDEX
    Explanations

    spare

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of personal free‐time (leisure) phrases, especially “spare time” contexts.

    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
     помощ
    -0.07
     Host
    -0.07
     pubs
    -0.07
     حض
    -0.06
     멀
    -0.06
     worsh
    -0.06
     นาง
    -0.06
    ogen
    -0.06
     medications
    -0.06
    _WR
    -0.06
    POSITIVE LOGITS
     віль
    0.07
    Sent
    0.07
     _.
    0.07
    loating
    0.06
    .cbo
    0.06
     Sanity
    0.06
    +(
    0.06
     spare
    0.06
     Spare
    0.06
    .↵↵↵↵
    0.06
    Activations Density 0.008%

    No Known Activations