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. 40477
    Prev
    Next
    INDEX
    Explanations

    local time

    np_max-act · gemini-2.0-flash

    issues relating to time zones and datetime conversions.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron fires on mentions of “time” and “timezone” (e.g. tokens relating to time-zones or time fields).

    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
    shirt
    -0.07
    irim
    -0.07
    }},↵
    -0.07
     "").
    -0.06
     Andrea
    -0.06
     MONTH
    -0.06
    ValueGenerationStrategy
    -0.06
     ",";↵
    -0.06
     acct
    -0.06
    crate
    -0.06
    POSITIVE LOGITS
     reclaimed
    0.07
    ilitating
    0.06
     electromagnetic
    0.06
     totalTime
    0.06
     climbing
    0.06
     Те
    0.06
    _usr
    0.06
    privileged
    0.06
    +y
    0.06
     ou
    0.06
    Activations Density 0.002%

    No Known Activations