© 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 EvalsExportsinterp-engineNEWAPI 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. 8-RES-JB
    11. 11345
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to government officials and organizations

    oai_token-act-pair · gpt-3.5-turbo

    references to senior officials and their roles within various organizations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Horus
    -0.83
    ynthesis
    -0.73
     TAMADRA
    -0.73
     Carnage
    -0.71
     Gors
    -0.71
    oooo
    -0.65
     Fukushima
    -0.65
    oooooooo
    -0.62
     Loki
    -0.62
    illions
    -0.61
    POSITIVE LOGITS
     analyst
    1.00
     wat
    1.00
     adviser
    0.99
     scholar
    0.96
     insider
    0.94
     negotiator
    0.93
     observer
    0.93
     researcher
    0.93
     colleague
    0.92
     ally
    0.92
    Activations Density 0.259%

    No Known Activations