© 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. 44223
    Prev
    Next
    INDEX
    Explanations

    tw

    np_max-act · gemini-2.0-flash

    The neuron is tuned to detect sequences of consecutive uppercase letters (letter n-grams) as appear in cipher texts or abbreviations.

    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
     статус
    -0.07
    ิดต
    -0.07
    命周期
    -0.06
     VLAN
    -0.06
    Pos
    -0.06
    Profiles
    -0.06
     consciousness
    -0.06
     floor
    -0.06
    METHOD
    -0.06
     ấm
    -0.06
    POSITIVE LOGITS
     depr
    0.07
    acing
    0.06
    acích
    0.06
     salvation
    0.06
     jackets
    0.06
     BJ
    0.06
    는
    0.06
    벤
    0.06
    �藏
    0.06
    reate
    0.06
    Activations Density 0.016%

    No Known Activations