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. 126829
    Prev
    Next
    INDEX
    Explanations

    methicillin

    np_max-act · gemini-2.0-flash

    This neuron activates on mentions of antibiotics with the “-cillin” suffix (e.g. methicillin, penicillin).

    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
    атів
    -0.06
    	read
    -0.06
    	k
    -0.06
    =
    -0.06
    響
    -0.06
    ArgsConstructor
    -0.06
     říká
    -0.06
    	m
    -0.06
    _No
    -0.06
     Dragon
    -0.06
    POSITIVE LOGITS
    illa
    0.07
    intersect
    0.07
    arra
    0.07
    ila
    0.06
    elli
    0.06
    nila
    0.06
    stit
    0.06
    configure
    0.06
    ecd
    0.06
    idal
    0.06
    Activations Density 0.001%

    No Known Activations