© 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. 41315
    Prev
    Next
    INDEX
    Explanations

    DIY projects/design blog snippets

    np_max-act · gemini-2.0-flash

    The neuron responds to very high‐frequency “glue” words and other common tokens (e.g. articles like “the,” demonstratives like “this,” conjunctions, pronouns, etc.). It’s basically a stop-word detector.

    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
    sey
    -0.06
     Sey
    -0.06
     Steve
    -0.06
     unknown
    -0.06
    zend
    -0.05
     Entries
    -0.05
    OrElse
    -0.05
    兄
    -0.05
    ま
    -0.05
    ETER
    -0.05
    POSITIVE LOGITS
    ahas
    0.07
    änder
    0.07
    anchor
    0.07
     juga
    0.07
     зал
    0.07
    :@"
    0.07
    //(
    0.07
     destruct
    0.07
    touches
    0.07
    rací
    0.06
    Activations Density 0.028%

    No Known Activations