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. 54865
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash

    The neuron fires on causal linking phrases, especially “as a result of” (and similar “result of” constructions).

    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
    IB
    -0.07
     Select
    -0.07
     familiar
    -0.06
     bucket
    -0.06
    hyth
    -0.06
     sits
    -0.06
     Microwave
    -0.06
     Charging
    -0.06
    �
    -0.06
     attracts
    -0.06
    POSITIVE LOGITS
    ρηση
    0.08
     lex
    0.07
    щення
    0.07
    번
    0.07
    xFD
    0.06
    洗
    0.06
     eventData
    0.06
    .scrollTop
    0.06
    ết
    0.06
    entlich
    0.06
    Activations Density 0.018%

    No Known Activations