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. 125487
    Prev
    Next
    INDEX
    Explanations

    Mechs and suits

    np_max-act · gemini-2.0-flash

    This neuron activates on mentions of “mech” (especially in “mech pilot”), i.e. detecting references to mecha.

    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
    Lin
    -0.06
     reap
    -0.06
    .blue
    -0.06
    	cb
    -0.06
    壊
    -0.06
    "Now
    -0.06
     příst
    -0.06
    .discount
    -0.06
    WidthSpace
    -0.06
     Wan
    -0.06
    POSITIVE LOGITS
     rendition
    0.07
    liers
    0.07
     boasting
    0.07
     MMA
    0.06
     belief
    0.06
     Flex
    0.06
     venues
    0.06
    .Tensor
    0.06
    -packed
    0.06
     lineman
    0.06
    Activations Density 0.016%

    No Known Activations