© 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. 68949
    Prev
    Next
    INDEX
    Explanations

    technical/legal documents

    np_max-act · gemini-2.0-flash

    This neuron detects the phrase “factually consistent,” i.e. language indicating an evaluation of factual consistency.

    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
     bringen
    -0.07
    �
    -0.07
    Overrides
    -0.06
    ський
    -0.06
     โดย
    -0.06
    -pic
    -0.06
     pierced
    -0.06
     Salv
    -0.06
     accompanied
    -0.06
    pll
    -0.06
    POSITIVE LOGITS
    hub
    0.07
     anth
    0.07
    子
    0.06
    venient
    0.06
     điều
    0.06
    .Images
    0.06
    84
    0.06
    _countries
    0.06
    |[
    0.06
    交流
    0.06
    Activations Density 0.001%

    No Known Activations