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. 43720
    Prev
    Next
    INDEX
    Explanations

    Stopwords

    np_max-act · gemini-2.0-flash

    The neuron fires on speculative statements about a potential first-time outcome—e.g. phrases like “could be the first victory.”

    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
     planes
    -0.07
     negatively
    -0.06
     concession
    -0.06
    .Escape
    -0.06
    530
    -0.06
    VertexUvs
    -0.06
     Fog
    -0.06
     Mall
    -0.06
    placed
    -0.06
    repo
    -0.06
    POSITIVE LOGITS
     electrom
    0.06
    雅黑
    0.06
     regul
    0.06
     사업
    0.06
    reaction
    0.06
     tộc
    0.06
     topical
    0.06
    =======
    0.06
     Azure
    0.06
    GRESS
    0.06
    Activations Density 0.283%

    No Known Activations