© 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. 8608
    Prev
    Next
    INDEX
    Explanations

    word or phrases related to political camps or factions

    oai_token-act-pair · gpt-3.5-turbo

    references to political parties and their actions

    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
    ixt
    -0.73
    uner
    -0.72
     Magnum
    -0.70
     solitude
    -0.69
    thood
    -0.68
    Lear
    -0.66
     LIFE
    -0.65
     emergencies
    -0.64
     age
    -0.62
     Seah
    -0.62
    POSITIVE LOGITS
     accuse
    1.26
     argue
    1.21
     vehemently
    1.21
     filib
    1.20
     insist
    1.20
     rallied
    1.12
     allege
    1.10
     vigorously
    1.10
     denounced
    1.09
     lobbied
    1.09
    Activations Density 0.364%

    No Known Activations