© 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. 43313
    Prev
    Next
    INDEX
    Explanations

    or

    np_max-act · gemini-2.0-flash

    This neuron fires when it encounters the conjunction “or” used to link alternate transport modes (e.g. “or boat”).

    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
    proved
    -0.07
     комп
    -0.07
    subscription
    -0.06
    िच
    -0.06
     aborted
    -0.06
     роз
    -0.06
    ^{-
    -0.06
    ünkü
    -0.06
     prove
    -0.06
    igest
    -0.06
    POSITIVE LOGITS
    (groups
    0.06
    .groups
    0.06
    Cooldown
    0.06
    �
    0.06
     prolong
    0.06
    ]--;↵
    0.06
     ロ
    0.06
     z
    0.06
     excer
    0.06
     등장
    0.06
    Activations Density 0.066%

    No Known Activations