© 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. 10-RES-JB
    11. 5485
    Prev
    Next
    INDEX
    Explanations

    words about authority, control, and action

    oai_token-act-pair · gpt-3.5-turbo

    connections between people and their interactions with authority or societal structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    was
    -0.95
     WAS
    -0.75
    MpServer
    -0.75
    Was
    -0.74
    been
    -0.74
    Had
    -0.73
     Been
    -0.72
     Was
    -0.71
    tained
    -0.70
    IFIED
    -0.68
    POSITIVE LOGITS
     decide
    1.25
     discover
    1.20
     manage
    1.19
     arrive
    1.16
     fail
    1.15
     stumble
    1.15
     strive
    1.14
     lose
    1.14
     announce
    1.13
     begin
    1.13
    Activations Density 0.518%

    No Known Activations