© 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. 7654
    Prev
    Next
    INDEX
    Explanations

    the verb "is" in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    repetitive structures, specifically the use of the phrase "is."

    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
    farious
    -0.58
    awed
    -0.56
    irlf
    -0.51
    aughed
    -0.50
    prise
    -0.47
    arij
    -0.46
    undred
    -0.46
    prising
    -0.45
    attering
    -0.45
    aterasu
    -0.45
    POSITIVE LOGITS
     is
    2.12
     isn
    1.67
     has
    1.58
     appears
    1.50
     does
    1.50
     was
    1.47
     tends
    1.46
     seems
    1.45
     doesn
    1.45
     becomes
    1.43
    Activations Density 0.536%

    No Known Activations