Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 27252
    Prev
    Next
    INDEX
    Explanations

    count

    np_max-act · gemini-2.0-flash

    This neuron detects emphatic, authority-marking keywords (like “statement,” “message,” “business,” “ask,” “good”) that appear at the end of sentences to underscore a point.

    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
     ale
    -0.07
     pym
    -0.07
    ANE
    -0.06
     tyto
    -0.06
    validator
    -0.06
    Mexico
    -0.06
    axe
    -0.06
     verbal
    -0.06
    iore
    -0.06
     blatantly
    -0.06
    POSITIVE LOGITS
    нос
    0.07
     appropriation
    0.07
     활동
    0.06
    ез
    0.06
     Marketable
    0.06
     doldur
    0.06
    _DICT
    0.06
    _FOCUS
    0.06
    geber
    0.06
    %c
    0.06
    Activations Density 0.381%

    No Known Activations