© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 85769
    Prev
    Next
    INDEX
    Explanations

    technical/engineering contexts

    np_max-act · gemini-2.0-flash

    technical terminologies related to electrical and electronic components.

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

    The neuron consistently activates on key technical or device-related nouns (e.g. apparatus, photon, glyph, S5) that identify the main subject or component in a formal description.

    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
     iterations
    -0.07
    dirs
    -0.06
    frared
    -0.06
     escap
    -0.06
    escape
    -0.06
    White
    -0.06
    abad
    -0.06
    Red
    -0.06
     плани
    -0.06
     beb
    -0.06
    POSITIVE LOGITS
     찬
    0.07
    (grammar
    0.07
    سبة
    0.06
    احل
    0.06
    الى
    0.06
     Blasio
    0.06
    :'',↵
    0.06
     ");↵↵
    0.06
     Sanct
    0.06
     assum
    0.06
    Activations Density 0.127%

    No Known Activations