© 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. 3-RES-JB
    11. 15353
    Prev
    Next
    INDEX
    Explanations

    mentions of celestial bodies, especially the moon

    oai_token-act-pair · gpt-3.5-turbo

    references to the lunar surface and associated celestial topics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    iveness
    -0.90
    lessly
    -0.86
    eva
    -0.85
    inelli
    -0.84
    Downloadha
    -0.84
    Regular
    -0.81
    éĹĺ
    -0.79
    resso
    -0.77
    illance
    -0.77
    holes
    -0.76
    POSITIVE LOGITS
     lunar
    1.34
     eclipse
    1.33
     Lunar
    1.11
     rover
    1.08
     orbit
    0.94
     orbits
    0.89
     moon
    0.89
     footprints
    0.87
     eclips
    0.85
     spacecraft
    0.84
    Activations Density 0.006%

    No Known Activations