© 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. 15780
    Prev
    Next
    INDEX
    Explanations

    terms related to specific places or organizations, such as "Rune," "Guthrie," "Geo Trust," and "Folk."

    oai_token-act-pair · gpt-3.5-turbo

    specific game titles or names related to gaming content

    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
     DSL
    -0.82
    andestine
    -0.81
     MIL
    -0.80
    orgetown
    -0.80
    Mil
    -0.75
     cov
    -0.75
     McCull
    -0.74
    CM
    -0.74
    Madison
    -0.74
     moms
    -0.73
    POSITIVE LOGITS
     Rune
    3.65
     Runes
    2.21
     rune
    2.03
     runes
    1.85
     Guth
    1.54
     Eid
    1.25
     Folk
    1.19
     Geo
    1.17
     Gi
    1.17
     Orb
    1.12
    Activations Density 0.050%

    No Known Activations