© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 3-RES-JB
    11. 4792
    Prev
    Next
    INDEX
    Explanations

    references to government-related matters, such as classified information, government records, and communication with government officials

    oai_token-act-pair · gpt-3.5-turbo

    references to government activities and classified information

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ahime
    -0.88
     Rider
    -0.75
     ç¥ŀ
    -0.73
     âĿ
    -0.71
     Survive
    -0.70
    imar
    -0.70
    aneers
    -0.69
     relegation
    -0.68
     :(
    -0.68
    shr
    -0.67
    POSITIVE LOGITS
    FBI
    1.38
     subpoen
    1.22
     FBI
    1.21
     intelligence
    1.21
    intelligence
    1.14
     investigative
    1.05
     vetting
    1.04
     dossier
    1.04
    afort
    1.03
     wiret
    1.02
    Activations Density 0.876%

    No Known Activations