© 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. 0-RES-JB
    11. 18275
    Prev
    Next
    INDEX
    Explanations

    references to legal or governmental powers

    oai_token-act-pair · gpt-3.5-turbo

    references to "Powers" in various contexts, indicating a focus on authority or legislative matters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     tube
    -0.69
     neck
    -0.65
     processed
    -0.63
     tolerance
    -0.62
     sugar
    -0.62
    UV
    -0.62
     tart
    -0.62
    icable
    -0.61
    cale
    -0.61
     coffin
    -0.60
    POSITIVE LOGITS
     Powers
    4.25
    powers
    1.85
     powers
    1.65
    owers
    1.22
     Power
    1.20
    Power
    1.12
     Patton
    1.10
     Commands
    1.09
     Allies
    1.04
     Nichols
    1.02
    Activations Density 0.017%

    No Known Activations