© 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
    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. 42565
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash

    This neuron fires on instances of the preposition “in” when it introduces a location or source (e.g. “in the file…,” “in the section…”).

    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
     søger
    -0.07
    :def
    -0.07
    .kr
    -0.07
     terk
    -0.06
     elde
    -0.06
    @store
    -0.06
    .dt
    -0.06
     تصم
    -0.06
    _DT
    -0.06
     starch
    -0.06
    POSITIVE LOGITS
    لة
    0.07
     '>'
    0.06
     pinpoint
    0.06
    EmailAddress
    0.06
     amalg
    0.06
     стать
    0.06
    MRI
    0.06
    startsWith
    0.06
    Stars
    0.06
     loyal
    0.06
    Activations Density 0.142%

    No Known Activations