© 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. 2-RES-JB
    11. 18684
    Prev
    Next
    INDEX
    Explanations

    words related to physical actions or physical characteristics

    oai_token-act-pair · gpt-3.5-turbo

    suffixes and endings of words, particularly gerunds and plural forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     unfor
    -0.46
     Shutterstock
    -0.44
     rapidly
    -0.43
     tightly
    -0.43
     commod
    -0.43
     prof
    -0.42
     nicely
    -0.42
     merciless
    -0.42
     cracked
    -0.42
     heavily
    -0.41
    POSITIVE LOGITS
    iversary
    0.69
    opolis
    0.63
    oking
    0.61
    arious
    0.61
    inite
    0.60
    verage
    0.60
    ogether
    0.60
    ulous
    0.59
    arius
    0.59
    tenance
    0.59
    Activations Density 1.508%

    No Known Activations