© 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. 8-RES-JB
    11. 20463
    Prev
    Next
    INDEX
    Explanations

    phrases related to numeric comparisons or multiples

    oai_token-act-pair · gpt-3.5-turbo

    comparative phrases involving numerical values and ratios

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    PLA
    -0.81
    azel
    -0.74
    corn
    -0.69
     Leilan
    -0.68
    behind
    -0.68
    fy
    -0.67
    ama
    -0.64
    ongo
    -0.64
    nothing
    -0.63
    RAW
    -0.63
    POSITIVE LOGITS
     amount
    1.21
     size
    1.06
     thickness
    1.04
     usual
    1.03
     extent
    1.01
     average
    0.98
     diameter
    0.97
     height
    0.88
     volume
    0.86
     magnitude
    0.86
    Activations Density 0.063%

    No Known Activations