© 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. 10314
    Prev
    Next
    INDEX
    Explanations

    mentions of companies

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of the word "company."

    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
    wards
    -0.66
    olkien
    -0.63
    ypes
    -0.62
    each
    -0.62
    lest
    -0.60
    ilib
    -0.60
    inct
    -0.60
    ophobia
    -0.59
    lected
    -0.58
    imore
    -0.58
    POSITIVE LOGITS
     company
    3.57
    company
    2.73
    Company
    2.60
     Company
    2.48
     companies
    2.29
    Companies
    2.17
     Companies
    2.05
     corporation
    2.03
     compan
    1.98
     firm
    1.97
    Activations Density 0.036%

    No Known Activations