© 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. 4853
    Prev
    Next
    INDEX
    Explanations

    the word "ki" with varying activation values

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of a specific name or concept

    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
     suppress
    -0.75
     distressed
    -0.69
     Sebast
    -0.67
     suppressing
    -0.66
     nond
    -0.64
     product
    -0.64
     suppression
    -0.61
     miracle
    -0.61
     Measure
    -0.60
     vacuum
    -0.60
    POSITIVE LOGITS
    ki
    4.32
    ky
    2.15
    ka
    1.98
    kaya
    1.86
    KI
    1.82
    kie
    1.74
    ku
    1.72
    ko
    1.71
    kies
    1.58
    ken
    1.54
    Activations Density 0.010%

    No Known Activations