© 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. 10-RES-JB
    11. 2254
    Prev
    Next
    INDEX
    Explanations

    words related to challenging or being challenged

    oai_token-act-pair · gpt-3.5-turbo

    contexts related to challenging authority or established norms

    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
    opter
    -0.68
    ãĥ¼ãĥĨãĤ£
    -0.68
    abet
    -0.65
     istg
    -0.65
    psons
    -0.64
    ppa
    -0.64
    ··
    -0.62
    anuts
    -0.61
    ]}
    -0.61
    around
    -0.61
    POSITIVE LOGITS
     assumptions
    1.04
     incumb
    0.93
     incumbent
    0.92
     precon
    0.90
     stereotypes
    0.89
     perceptions
    0.87
     orthodoxy
    0.86
     misconceptions
    0.84
     boundaries
    0.78
     assertions
    0.78
    Activations Density 0.066%

    No Known Activations