© 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. 128539
    Prev
    Next
    INDEX
    Explanations

    assumption

    np_max-act · gemini-2.0-flash

    The neuron specializes in spotting discussion-style methodological terms (like “assumption,” “simplifying,” “analysis,” “appropriate,” “context,” etc.) that evaluate or qualify the approach.

    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
    γκ
    -0.07
     Wol
    -0.06
     Lim
    -0.06
    Trou
    -0.06
    ικα
    -0.06
     Altern
    -0.06
    ');
↵
    -0.06
    _Dep
    -0.06
     vocal
    -0.06
     Bull
    -0.06
    POSITIVE LOGITS
    莎
    0.07
    otu
    0.07
     renk
    0.07
    rouw
    0.07
     水
    0.06
     ترك
    0.06
     içine
    0.06
     corro
    0.06
     tisk
    0.06
     Ub
    0.06
    Activations Density 0.028%

    No Known Activations