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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 55147
    Prev
    Next
    INDEX
    Explanations

    assured events

    np_max-act · gemini-2.0-flash

    the neuron detects verbs that express securing or guaranteeing an outcome (e.g., guaranteed, secured, clinched, locked).

    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
    .ParseException
    -0.07
    quer
    -0.06
    .getFirst
    -0.06
     Grape
    -0.06
    Mirror
    -0.06
    ++){↵↵
    -0.06
    NonQuery
    -0.06
    ása
    -0.06
     Нет
    -0.06
     sarà
    -0.06
    POSITIVE LOGITS
     århus
    0.06
    .Tr
    0.06
    reed
    0.06
    -linux
    0.06
    ROTO
    0.06
     REGARD
    0.06
    UIL
    0.06
    بع
    0.06
     infiltr
    0.06
     köln
    0.06
    Activations Density 0.043%

    No Known Activations