© 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. 14555
    Prev
    Next
    INDEX
    Explanations

    terms related to dedication or commitment to a cause or belief

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "voted" and its variations in the context of political discussions

    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
    ptoms
    -0.73
    pton
    -0.66
    atha
    -0.64
    Phys
    -0.63
    flix
    -0.62
     Marble
    -0.59
    session
    -0.58
    Sullivan
    -0.58
    Split
    -0.58
     dissolve
    -0.57
    POSITIVE LOGITS
    oted
    1.16
    inel
    0.92
    otes
    0.79
    agues
    0.77
    ragon
    0.75
    imentary
    0.75
    leck
    0.75
    oting
    0.75
    lasses
    0.72
    nikov
    0.71
    Activations Density 0.005%

    No Known Activations