© 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. 9448
    Prev
    Next
    INDEX
    Explanations

    terms related to higher education

    oai_token-act-pair · gpt-3.5-turbo

    references to higher education and election-related topics

    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
     captcha
    -0.75
     butt
    -0.74
     backup
    -0.74
     attrition
    -0.72
     perfection
    -0.70
     stockp
    -0.70
     spare
    -0.69
     pie
    -0.69
     knockout
    -0.68
     infect
    -0.68
    POSITIVE LOGITS
     Higher
    2.46
     Elections
    2.09
     Internal
    1.74
     Trad
    1.66
    Higher
    1.65
     Equal
    1.53
     Housing
    1.41
     Religious
    1.32
     Controlled
    1.30
     Industrial
    1.30
    Activations Density 0.049%

    No Known Activations