© 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. 44364
    Prev
    Next
    INDEX
    Explanations

    value

    np_max-act · gemini-2.0-flash

    The neuron is tuned to detect decimal‐formatted numerical values (e.g. performance metrics like “0.4128,” “2.6816,” “4.0820,” etc.).

    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
     IMAGES
    -0.07
     }</
    -0.07
    "/>.</
    -0.06
    }/>↵
    -0.06
    oters
    -0.06
     scape
    -0.06
     delight
    -0.06
    developers
    -0.06
    .Warning
    -0.06
     clever
    -0.06
    POSITIVE LOGITS
    اسي
    0.07
     daar
    0.07
    andFilterWhere
    0.07
    emente
    0.07
     wherever
    0.06
     containing
    0.06
    ji
    0.06
     initiatives
    0.06
     enjoying
    0.06
    νε
    0.06
    Activations Density 0.003%

    No Known Activations