© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 109882
    Prev
    Next
    INDEX
    Explanations

    by

    np_max-act · gemini-2.0-flash

    This neuron detects the word “by” when it introduces an explanatory mechanism (e.g. “works by…”, “reducing by…”) in technical descriptions.

    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
    TOCOL
    -0.06
     вік
    -0.06
     chút
    -0.06
    ають
    -0.06
     эта
    -0.06
     vrouwen
    -0.06
    usions
    -0.06
     Xen
    -0.06
    Fs
    -0.05
    _PE
    -0.05
    POSITIVE LOGITS
    _CD
    0.07
     subtitle
    0.07
     olm
    0.06
     Promise
    0.06
    promise
    0.06
    决
    0.06
    innie
    0.06
     Louisville
    0.06
    .ibm
    0.06
     नद
    0.06
    Activations Density 0.008%

    No Known Activations