© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 6307
    Prev
    Next
    INDEX
    Explanations

    mentions of processes that improve or enhance something

    oai_token-act-pair · gpt-3.5-turbo

    terminology related to enrichment, particularly in the context of resources or materials

    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
     dispatch
    -0.72
     Tele
    -0.70
     snapped
    -0.68
     Deadpool
    -0.68
     Cable
    -0.68
     Rider
    -0.67
     rookie
    -0.64
    head
    -0.64
     Hawth
    -0.64
    Tele
    -0.63
    POSITIVE LOGITS
     enrichment
    3.52
     enrich
    3.51
     enriched
    3.50
    riched
    1.82
     centrif
    1.47
     impover
    1.44
     richer
    1.27
     uranium
    1.25
     richness
    1.22
    utonium
    1.00
    Activations Density 0.029%

    No Known Activations