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. 95
    Prev
    Next
    INDEX
    Explanations

    can

    np_max-act · gemini-2.0-flash

    The neuron fires strongly on the verb “used” when it’s immediately followed by “to,” i.e. it picks out occurrences of the phrase “used to.”

    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
    监听
    -0.07
    justify
    -0.07
     Registrar
    -0.07
     tuổi
    -0.07
     hack
    -0.07
    .Version
    -0.06
    LU
    -0.06
     mods
    -0.06
    .Utils
    -0.06
    Ground
    -0.06
    POSITIVE LOGITS
    teg
    0.07
     perpetrated
    0.06
    Nested
    0.06
    .SerializeObject
    0.06
    864
    0.06
    ěj
    0.06
    945
    0.06
     именно
    0.06
    �
    0.06
    acies
    0.06
    Activations Density 0.522%

    No Known Activations