© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 16189
    Prev
    Next
    INDEX
    Explanations

    references to political figures, especially related to the office of Secretary of State

    oai_token-act-pair · gpt-3.5-turbo

    references to "Secretary of State," particularly Hillary Clinton

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    wcs
    -0.77
    culosis
    -0.72
    anchez
    -0.71
    nces
    -0.71
     dots
    -0.70
    oother
    -0.68
    iquid
    -0.67
    akeru
    -0.66
     flashbacks
    -0.64
     seams
    -0.64
    POSITIVE LOGITS
     Homeland
    0.73
     Communications
    0.71
    ghazi
    0.70
    Correction
    0.69
     course
    0.69
    Initialized
    0.68
     Instruction
    0.68
     presidential
    0.68
    µ
    0.68
     Communication
    0.67
    Activations Density 0.076%

    No Known Activations