© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 91649
    Prev
    Next
    INDEX
    Explanations

    operation

    np_max-act · gemini-2.0-flash

    This neuron detects quantitative result mentions—tokens introducing numerical measurements or metrics (e.g. operation times, follow-up durations, p-values) in clinical study reports.

    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
     entertainment
    -0.08
     calf
    -0.06
     leased
    -0.06
    _cards
    -0.06
    (matches
    -0.06
     dissemination
    -0.06
    安装
    -0.06
     shoppers
    -0.06
     homelessness
    -0.06
    $a
    -0.06
    POSITIVE LOGITS
     lavor
    0.07
    ellig
    0.07
    .LogError
    0.07
     seksi
    0.07
     inan
    0.07
     مرتبط
    0.07
    .DateTimePicker
    0.07
    >,↵
    0.07
    ере
    0.07
     MOD
    0.06
    Activations Density 0.008%

    No Known Activations