© 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. 857
    Prev
    Next
    INDEX
    Explanations

    TV listings

    np_max-act · gemini-2.0-flash

    This neuron responds to tokens containing numbers (e.g. timestamps, ratings, measurements, or other digit‐based values).

    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
     future
    -0.07
    877
    -0.07
     pá
    -0.07
     fmt
    -0.07
     MessageBoxIcon
    -0.06
     身
    -0.06
    .admin
    -0.06
     homework
    -0.06
     unreliable
    -0.06
     Thủ
    -0.06
    POSITIVE LOGITS
     Česko
    0.07
    -coded
    0.07
    	al
    0.06
    live
    0.06
    (options
    0.06
    Opts
    0.06
    setColor
    0.06
     případ
    0.06
    grily
    0.06
     lizard
    0.06
    Activations Density 0.008%

    No Known Activations