© 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. 8840
    Prev
    Next
    INDEX
    Explanations

    chemistry

    np_max-act · gemini-2.0-flash

    The neuron activates on terms that describe chemical‐synthesis steps—especially mentioning reagents, reaction conditions, and process verbs (e.g. “in the presence of,” “using,” “synthesized by,” reagent names, catalysts).

    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
    .notification
    -0.07
    $start
    -0.06
    ような
    -0.06
     Tracking
    -0.06
     question
    -0.06
    ysics
    -0.06
    Чтобы
    -0.06
     Pyramid
    -0.06
    ール
    -0.06
     filles
    -0.06
    POSITIVE LOGITS
    �
    0.07
    .SwingConstants
    0.07
    owej
    0.07
    (Edit
    0.07
    　ヾ
    0.07
     Hopefully
    0.07
     엄
    0.07
    ]",
    0.06
    secured
    0.06
    BootApplication
    0.06
    Activations Density 0.054%

    No Known Activations