© 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. 13558
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash

    This neuron identifies mentions of the admissibility of evidence (e.g. “admissibility of expert testimony”) in legal 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
     STDMETHOD
    -0.07
    τρο
    -0.07
    mental
    -0.07
    ioc
    -0.07
     defs
    -0.07
     offices
    -0.06
    šen
    -0.06
    ab
    -0.06
     vita
    -0.06
    куль
    -0.06
    POSITIVE LOGITS
     Exists
    0.07
    [href
    0.06
    raries
    0.06
    -final
    0.06
    CLUDED
    0.05
     한다
    0.05
     neob
    0.05
    !(:
    0.05
     조금
    0.05
    23
    0.05
    Activations Density 0.011%

    No Known Activations