© 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. 7-RES-JB
    11. 1463
    Prev
    Next
    INDEX
    Explanations

    the word "to" followed by another word or a number, potentially related to expectations or obligations

    oai_token-act-pair · gpt-3.5-turbo

    expectations or duties related to events or situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Appears
    -0.71
     Compass
    -0.68
     Tens
    -0.67
     quotations
    -0.66
    iple
    -0.63
     Levi
    -0.62
     Surge
    -0.61
    river
    -0.61
     conceptions
    -0.60
     Kis
    -0.58
    POSITIVE LOGITS
     be
    0.98
     behave
    0.94
     solve
    0.92
     protect
    0.91
     embody
    0.88
     represent
    0.88
     compensate
    0.88
     equate
    0.87
     adhere
    0.86
     defend
    0.86
    Activations Density 0.081%

    No Known Activations