© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 29717
    Prev
    Next
    INDEX
    Explanations

    Research/Studies

    np_max-act · gemini-2.0-flash

    The neuron activates on nouns naming key evaluative metrics or constructs in academic reports (e.g. recovery, vulnerability, outcome, empowerment, compliance, emergence, burden).

    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
    规划
    -0.07
     dataSource
    -0.06
    Credential
    -0.06
    897
    -0.06
    és
    -0.06
     ballistic
    -0.06
    Hour
    -0.06
    /W
    -0.06
    ิน
    -0.06
    เท
    -0.06
    POSITIVE LOGITS
    _direction
    0.07
     سخ
    0.07
     isAuthenticated
    0.06
     vyžad
    0.06
     casa
    0.06
     accountable
    0.06
    );*/↵
    0.06
     medios
    0.06
     ironically
    0.06
    .Margin
    0.06
    Activations Density 0.052%

    No Known Activations