© 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. 126783
    Prev
    Next
    INDEX
    Explanations

    speech or thought

    np_max-act · gemini-2.0-flash

    information related to business growth and financial investments.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron responds to occurrences of the verb “say,” especially in phrases like “People say” that introduce reported opinions.

    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
     seeding
    -0.06
    _estimators
    -0.06
    uni
    -0.06
    儿
    -0.06
    Cu
    -0.06
    işti
    -0.06
    _unsigned
    -0.06
     nuevo
    -0.06
     Pun
    -0.06
    _read
    -0.06
    POSITIVE LOGITS
     preval
    0.07
     "','
    0.07
    ▋
    0.07
    "]."
    0.07
     khúc
    0.07
    ینه
    0.07
    ">'
    0.07
     전용면적
    0.06
    Supply
    0.06
    ;\
    0.06
    Activations Density 0.029%

    No Known Activations