© 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. 22536
    Prev
    Next
    INDEX
    Explanations

    mentions of words containing the substring "CO"

    oai_token-act-pair · gpt-3.5-turbo

    references to the CO abbreviation, likely related to carbon dioxide or similar scientific contexts

    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
     Shant
    -0.73
     Flesh
    -0.67
     emanc
    -0.67
     Shiv
    -0.66
     Yosh
    -0.65
     lich
    -0.65
     Shy
    -0.64
     rash
    -0.64
     Quest
    -0.64
     slaying
    -0.63
    POSITIVE LOGITS
    CO
    4.23
    co
    1.84
     CO
    1.76
    CA
    1.71
    COM
    1.64
    Co
    1.54
    SO
    1.50
    MO
    1.49
    PO
    1.42
    CON
    1.37
    Activations Density 0.007%

    No Known Activations