© 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. 4-RES-JB
    11. 7093
    Prev
    Next
    INDEX
    Explanations

    mentions of individuals named "Sam" or "Samuel" in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    repetitive references to the term "am."

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

    No Comments

    Negative Logits
     steroids
    -0.76
     stakes
    -0.72
     masks
    -0.67
     puberty
    -0.65
     takeaway
    -0.64
     tremend
    -0.64
     uranium
    -0.63
     resemb
    -0.62
     edges
    -0.62
     corners
    -0.62
    POSITIVE LOGITS
    endment
    1.29
    nesty
    1.27
    ilies
    1.13
    sterdam
    1.11
    bitious
    1.09
    borgh
    1.09
    essage
    1.09
    urai
    1.04
    azing
    1.02
    aha
    1.01
    Activations Density 0.023%

    No Known Activations