© 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. 8-RES-JB
    11. 22366
    Prev
    Next
    INDEX
    Explanations

    terms related to legal, political, or social issues

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to military context or operations

    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
    colour
    -0.87
     colours
    -0.87
     favourites
    -0.80
     fulfil
    -0.80
     colour
    -0.80
     armour
    -0.78
     humour
    -0.77
     apologise
    -0.77
     organise
    -0.77
     brill
    -0.77
    POSITIVE LOGITS
     HHS
    0.97
     bipartisan
    0.96
     Congressional
    0.91
     antitrust
    0.90
     federally
    0.90
     POLITICO
    0.90
     DOJ
    0.89
    Federal
    0.87
     ALEC
    0.86
     Politico
    0.85
    Activations Density 1.733%

    No Known Activations