© 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. 1-RES-JB
    11. 6598
    Prev
    Next
    INDEX
    Explanations

    mentions of strong emotions or extreme attitudes

    oai_token-act-pair · gpt-3.5-turbo

    words associated with extreme partisanship or fanatical behavior

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Simone
    -0.73
    neau
    -0.71
     Ell
    -0.71
     Tanaka
    -0.69
     Sisters
    -0.69
     Cout
    -0.68
     Pebble
    -0.68
     Sacrament
    -0.68
     Duncan
    -0.65
     Ernst
    -0.65
    POSITIVE LOGITS
     rab
    1.23
    ãĥ£
    0.95
    bats
    0.87
    inary
    0.83
    verning
    0.82
    pox
    0.79
    wcs
    0.79
    shit
    0.78
    pload
    0.78
     rabid
    0.77
    Activations Density 0.020%

    No Known Activations