© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 89640
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron fires primarily on numeric tokens—especially decimal or fractional values—within descriptive text.

    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
    _NEAREST
    -0.07
    ظمة
    -0.06
     #"
    -0.06
    _Map
    -0.06
    .ali
    -0.06
     flagship
    -0.06
     Valk
    -0.06
    .maps
    -0.06
    opy
    -0.06
    .Abs
    -0.06
    POSITIVE LOGITS
    İK
    0.07
     ORD
    0.06
     cr
    0.06
    데이트
    0.06
     anarchists
    0.06
    AAAAAAAA
    0.06
     спів
    0.06
     kur
    0.06
    _MINOR
    0.06
     chave
    0.06
    Activations Density 0.068%

    No Known Activations