© 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. 600
    Prev
    Next
    INDEX
    Explanations

    ex

    np_max-act · gemini-2.0-flash

    This neuron activates on tokens that start a sentence (i.e. sentence-initial, often capitalized, words).

    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
    perimental
    -0.06
    /platform
    -0.06
    justify
    -0.06
    .t
    -0.06
     smell
    -0.06
     ruler
    -0.06
    .b
    -0.06
     substitution
    -0.06
    adult
    -0.06
    EMU
    -0.05
    POSITIVE LOGITS
     sass
    0.07
    ンパ
    0.07
    िप
    0.07
     unanim
    0.07
    ⠀⠀
    0.06
     fidelity
    0.06
    ="#">↵
    0.06
     devez
    0.06
    /fixtures
    0.06
    μένου
    0.06
    Activations Density 0.437%

    No Known Activations