© 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. 91987
    Prev
    Next
    INDEX
    Explanations

    causation

    np_max-act · gemini-2.0-flash

    The neuron responds to auxiliary/modal or causative verbs (e.g. “can,” “makes,” “means,” “requires”) that introduce capabilities or causal relationships.

    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
     LEN
    -0.06
    Doc
    -0.06
     nhựa
    -0.06
     circulation
    -0.06
    _video
    -0.06
    etailed
    -0.06
     impressions
    -0.06
     method
    -0.06
    .pi
    -0.06
    SPATH
    -0.06
    POSITIVE LOGITS
    nex
    0.08
    _MIC
    0.07
     strstr
    0.07
    _remain
    0.06
     itemBuilder
    0.06
    _FWD
    0.06
    tests
    0.06
    ochrome
    0.06
     puede
    0.06
    _portfolio
    0.06
    Activations Density 0.071%

    No Known Activations