© 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. 0-RES-JB
    11. 1595
    Prev
    Next
    INDEX
    Explanations

    the word "of" used in different contexts

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "Of course."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     special
    -0.62
     ribbon
    -0.61
     close
    -0.61
     professionally
    -0.61
     once
    -0.58
     dead
    -0.58
     a
    -0.57
     fire
    -0.57
     internationally
    -0.57
     state
    -0.57
    POSITIVE LOGITS
    Of
    3.26
     Of
    1.92
    of
    1.65
    Naturally
    1.49
    OF
    1.47
    Between
    1.44
    Against
    1.42
    Indeed
    1.38
    oft
    1.38
    To
    1.35
    Activations Density 0.014%

    No Known Activations