© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 73781
    Prev
    Next
    INDEX
    Explanations

    large amounts

    np_max-act · gemini-2.0-flash

    The neuron detects technical qualifier words and phrases that specify conditions or intensify problem descriptions (e.g. “when,” “extremely,” “amplified,” “larger,” “wherein”).

    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
    Rare
    -0.07
    ْت
    -0.07
    َم
    -0.07
     sce
    -0.06
     tagging
    -0.06
    Crit
    -0.06
     imagem
    -0.06
    287
    -0.06
     hob
    -0.06
    Camb
    -0.06
    POSITIVE LOGITS
     Midi
    0.07
    $data
    0.07
    ROUND
    0.07
     Freeman
    0.07
    estination
    0.07
     vận
    0.06
    @c
    0.06
     recursive
    0.06
    sv
    0.06
    *******↵
    0.06
    Activations Density 0.055%

    No Known Activations