© 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. 129971
    Prev
    Next
    INDEX
    Explanations

    Arrows/Lines

    np_max-act · gemini-2.0-flash

    This neuron fires on numeric revision references and count tokens (e.g. “15,” “4,” “HEAD~4,” etc.) used to specify commits or offsets in version‐control operations.

    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
    .fold
    -0.06
    =message
    -0.06
    anya
    -0.06
    蜜
    -0.06
     Ergebn
    -0.06
     punched
    -0.05
     hoàng
    -0.05
     taught
    -0.05
     Bonus
    -0.05
    圖
    -0.05
    POSITIVE LOGITS
    summer
    0.07
     ava
    0.07
    condition
    0.07
    processing
    0.07
    Hopefully
    0.06
    CHANT
    0.06
     Συ
    0.06
     SM
    0.06
    idl
    0.06
    whole
    0.06
    Activations Density 0.007%

    No Known Activations