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. 16190
    Prev
    Next
    INDEX
    Explanations

    scientific/technical documents

    np_max-act · gemini-2.0-flash

    This neuron responds to patent-style listings of alternative technical embodiments, i.e. phrases enumerating different device or method variants (e.g. “a one using …,” “a one for displacing …,” “a one using means for generating …”).

    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
     ка
    -0.07
     мощ
    -0.07
    共同
    -0.06
    .heading
    -0.06
    .evaluate
    -0.06
    .t
    -0.06
    ']))
    -0.06
     penc
    -0.06
     thọ
    -0.06
    .Cancel
    -0.06
    POSITIVE LOGITS
    illy
    0.07
     Surge
    0.07
    μπο
    0.06
    .readdir
    0.06
     STACK
    0.06
    icut
    0.06
     fours
    0.06
     corpus
    0.06
    ifton
    0.06
     dumb
    0.06
    Activations Density 0.113%

    No Known Activations