© 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. 1-RES-JB
    11. 13197
    Prev
    Next
    INDEX
    Explanations

    company names ending with "Inc"

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the term "Inc" associated with companies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    BOOK
    -0.88
    */(
    -0.85
    =-=-=-=-=-=-=-=-
    -0.73
    =\"
    -0.69
    actionDate
    -0.69
    wic
    -0.68
    Redditor
    -0.67
    lled
    -0.66
    PsyNetMessage
    -0.65
    fulness
    -0.65
    POSITIVE LOGITS
    orporated
    1.64
    orpor
    1.20
    arcer
    1.13
    enture
    1.06
    orp
    1.00
    endiary
    1.00
    orrect
    0.93
    ursions
    0.89
     Inc
    0.89
     incorpor
    0.89
    Activations Density 0.005%

    No Known Activations