© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 68319
    Prev
    Next
    INDEX
    Explanations

    structure

    np_max-act · gemini-2.0-flash

    The neuron activates on technical descriptions of exploiting special algebraic or structural forms (e.g. “special form,” “structure,” “exploiting,” “optimization”) in algorithmic or mathematical contexts.

    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
     honoured
    -0.07
     grado
    -0.06
     YORK
    -0.06
     önce
    -0.06
     contributors
    -0.06
    _th
    -0.06
     pounds
    -0.06
     latent
    -0.06
     shameful
    -0.06
    -group
    -0.06
    POSITIVE LOGITS
    .userService
    0.07
    Fall
    0.07
    _matching
    0.06
    Override
    0.06
    	Json
    0.06
    .Result
    0.06
    -centric
    0.06
    _expand
    0.06
     empres
    0.06
    Templates
    0.06
    Activations Density 0.076%

    No Known Activations