© 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. 10421
    Prev
    Next
    INDEX
    Explanations

    numbers indicating quantity, specifically the number four

    oai_token-act-pair · gpt-3.5-turbo

    the word "four" and its numerical significance in various contexts

    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
    RL
    -0.67
    soType
    -0.65
    ================================================================
    -0.61
    liness
    -0.61
     disapp
    -0.59
    quit
    -0.58
     Allows
    -0.58
    ilo
    -0.57
    BuyableInstoreAndOnline
    -0.57
    qi
    -0.57
    POSITIVE LOGITS
     four
    2.81
     five
    2.37
     three
    2.36
     six
    2.30
     eight
    2.27
     seven
    2.20
    four
    2.12
     nine
    2.08
     two
    1.96
    Four
    1.90
    Activations Density 0.030%

    No Known Activations