© 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. 68817
    Prev
    Next
    INDEX
    Explanations

    Technical documentation

    np_max-act · gemini-2.0-flash

    The neuron flags the main technical object or domain‐specific noun (e.g. “xml file,” “classes,” “malocclusion,” etc.) that is the central subject of each question.

    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
    acked
    -0.06
    ピー
    -0.06
    mit
    -0.06
    то
    -0.06
    rts
    -0.06
    Mid
    -0.06
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    -0.06
    .sync
    -0.06
     graphene
    -0.06
     massages
    -0.06
    POSITIVE LOGITS
    lobals
    0.07
    _HIDDEN
    0.06
    -related
    0.06
    ­های
    0.06
     DISCLAIMS
    0.06
     connexion
    0.06
    	InputStream
    0.06
    reuse
    0.06
     anlam
    0.06
    -alert
    0.06
    Activations Density 0.172%

    No Known Activations