© 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. 21084
    Prev
    Next
    INDEX
    Explanations

    keywords related to planets and celestial bodies

    oai_token-act-pair · gpt-3.5-turbo

    references to planets

    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
    headers
    -0.80
    Repe
    -0.73
    RFC
    -0.66
     Marketable
    -0.65
    WER
    -0.64
    IME
    -0.64
    IFF
    -0.63
    ulton
    -0.63
     interpreter
    -0.63
     GROUP
    -0.61
    POSITIVE LOGITS
    arium
    1.54
     Earth
    1.23
     Venus
    1.21
    oid
    1.19
     orbiting
    1.13
     Ceres
    1.12
    ouched
    1.12
     earth
    1.07
     Mars
    0.99
     habitable
    0.99
    Activations Density 0.025%

    No Known Activations