© 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. 89347
    Prev
    Next
    INDEX
    Explanations

    N

    np_max-act · gemini-2.0-flash

    The neuron activates on fragments of time‐related field names (e.g. “Sec,” “Nsec,” “Usec,” “Actime,” “Modtime,” etc.) in generated Go struct definitions.

    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
     melhor
    -0.07
    Checkpoint
    -0.06
    .time
    -0.06
     Lund
    -0.06
    惠
    -0.06
     Historic
    -0.06
    adastrar
    -0.06
     setPage
    -0.06
     Dept
    -0.06
    ंटर
    -0.06
    POSITIVE LOGITS
    }".
    0.07
    ificate
    0.07
     Come
    0.06
    IV
    0.06
     prosec
    0.06
    uv
    0.06
    -complete
    0.06
     Drama
    0.06
    (rows
    0.06
    �
    0.06
    Activations Density 0.003%

    No Known Activations