© 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. 10-RES-JB
    11. 7744
    Prev
    Next
    INDEX
    Explanations

    phrases related to issues, trends, and patterns

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to significant social or systemic issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    endars
    -0.83
    inventoryQuantity
    -0.70
    emort
    -0.69
     Carbuncle
    -0.67
    姫
    -0.67
    ãĥ¼ãĥ³
    -0.67
    ãĥīãĥ©
    -0.66
     equivalents
    -0.64
     wills
    -0.63
     Tsukuyomi
    -0.63
    POSITIVE LOGITS
     endemic
    1.13
     plag
    1.09
     inherent
    1.01
     prevalent
    0.98
     whereby
    0.97
     perv
    0.96
     pervasive
    0.95
     rooted
    0.95
     underlying
    0.94
     occurring
    0.88
    Activations Density 0.306%

    No Known Activations