© 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. 20501
    Prev
    Next
    INDEX
    Explanations

    quantitative data points expressed as percentages

    oai_token-act-pair · gpt-3.5-turbo

    percentage values or statistical data in a text

    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
     exposition
    -0.70
     smugglers
    -0.67
    ĸļ
    -0.67
     feats
    -0.67
     successors
    -0.66
     cliff
    -0.63
     thorn
    -0.62
     flashbacks
    -0.62
     bombardment
    -0.62
     reinvent
    -0.60
    POSITIVE LOGITS
    percent
    1.05
    approximately
    1.03
    excluding
    1.03
    %)
    1.02
    56
    0.97
    46
    0.96
    76
    0.96
    57
    0.95
    74
    0.95
    55
    0.93
    Activations Density 0.088%

    No Known Activations