© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 11-RES-JB
    11. 16299
    Prev
    Next
    INDEX
    Explanations

    terms related to corruption and ethics in governmental and political contexts

    oai_token-act-pair · gpt-3.5-turbo

    references to corruption and its impact

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    leigh
    -0.75
    nesium
    -0.66
    imus
    -0.61
     Clockwork
    -0.61
    ertility
    -0.61
     childbirth
    -0.61
    keley
    -0.60
     pacif
    -0.60
     Gladiator
    -0.60
    ocard
    -0.59
    POSITIVE LOGITS
     scandals
    1.32
     allegations
    1.02
     endemic
    0.98
    ulent
    0.95
     scandal
    0.93
     investigations
    0.91
     accusations
    0.89
     rife
    0.89
     perpetrated
    0.88
     watchdog
    0.87
    Activations Density 0.094%

    No Known Activations