© 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. 55772
    Prev
    Next
    INDEX
    Explanations

    locations

    np_max-act · gemini-2.0-flash

    historical references and significant events related to individuals and places.

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

    The neuron activates on the relative adverb “where” (and its immediate context) when it introduces a location‐describing clause.

    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
    .Level
    -0.08
     soil
    -0.07
    Tuple
    -0.06
    (timer
    -0.06
    Ur
    -0.06
    Ed
    -0.06
    Sou
    -0.06
     structured
    -0.06
    ansion
    -0.06
     wound
    -0.06
    POSITIVE LOGITS
    ]+$
    0.07
     поє
    0.07
     "','"
    0.07
    "]="
    0.06
    }".
    0.06
    ).'</
    0.06
    거래가
    0.06
    .cursor
    0.06
     вд
    0.06
    .pid
    0.06
    Activations Density 0.040%

    No Known Activations