© 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. 18019
    Prev
    Next
    INDEX
    Explanations

    stuck or trapped

    np_max-act · gemini-2.0-flash

    phrases related to vehicle design and safety.

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

    The main thing this neuron does is detect language describing objects becoming physically stuck, caught, or otherwise jammed.

    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
     Ade
    -0.06
     gd
    -0.06
    _read
    -0.06
    .branch
    -0.06
    Cont
    -0.06
    Line
    -0.06
     unordered
    -0.06
    Promise
    -0.06
    л
    -0.06
    iê
    -0.06
    POSITIVE LOGITS
    eful
    0.06
    ун
    0.06
    /tinyos
    0.06
     �
    0.06
    -US
    0.06
     --------------------------------------------------------------------------↵
    0.06
    /commons
    0.06
     Indices
    0.06
     destinations
    0.06
    intptr
    0.06
    Activations Density 0.025%

    No Known Activations