© 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. 126027
    Prev
    Next
    INDEX
    Explanations

    dose

    np_max-act · gemini-2.0-flash

    phrases related to dosage and its effects in medical contexts.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on mentions of “dose” (and its close variants like “doses,” “dose-response,” or “dosing”).

    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
     Kern
    -0.07
    приєм
    -0.07
     rocky
    -0.07
     remembered
    -0.07
     kní
    -0.07
     občan
    -0.07
    children
    -0.06
     kitt
    -0.06
     pancakes
    -0.06
     McKay
    -0.06
    POSITIVE LOGITS
     dose
    0.15
     doses
    0.12
     dosage
    0.10
     overdose
    0.09
    _SE
    0.08
     Dos
    0.08
    Dos
    0.07
    ose
    0.07
    base
    0.07
    se
    0.07
    Activations Density 0.010%

    No Known Activations