© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 11-RES-JB
    11. 9292
    Prev
    Next
    INDEX
    Explanations

    percentage changes from a reference value

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating changes in percentages or statistics over time

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     proble
    -0.79
    isphere
    -0.68
    é¾įåĸļ士
    -0.66
    Reviewer
    -0.65
     conflic
    -0.65
    Others
    -0.62
    Winged
    -0.62
    AZ
    -0.60
     differ
    -0.60
    èª
    -0.58
    POSITIVE LOGITS
     yesterday
    1.01
     2005
    0.88
     2007
    0.86
     whence
    0.85
     2010
    0.85
     last
    0.85
     2009
    0.85
     2008
    0.84
     previous
    0.84
     1979
    0.83
    Activations Density 0.092%

    No Known Activations