© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 92168
    Prev
    Next
    INDEX
    Explanations

    commas/separators

    np_max-act · gemini-2.0-flash

    contextual information about pet ownership, particularly regarding dogs.

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

    The neuron is looking for words or tokens that signal non-English/foreign-language text (i.e. it spikes on words in Italian, Spanish, etc.).

    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
    ASET
    -0.08
     Given
    -0.06
    _beh
    -0.06
    _TESTS
    -0.06
     Gos
    -0.06
     given
    -0.06
    .SpringBootTest
    -0.06
     impr
    -0.06
     dar
    -0.06
    UIViewController
    -0.06
    POSITIVE LOGITS
     withRouter
    0.07
     breathable
    0.07
    creative
    0.07
    supplier
    0.06
    _tri
    0.06
     invis
    0.06
     ->
    0.06
    rott
    0.06
     femin
    0.06
    perms
    0.06
    Activations Density 0.056%

    No Known Activations