Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 15980
    Prev
    Next
    INDEX
    Explanations

    terms related to corporations or corporate activities

    oai_token-act-pair · gpt-3.5-turbo

    references to corporate entities and activities

    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
    emen
    -0.69
    ivari
    -0.68
     Ages
    -0.68
    Keys
    -0.67
    Redditor
    -0.66
    aceae
    -0.66
    Adds
    -0.65
    iev
    -0.65
    Range
    -0.65
     Eh
    -0.65
    POSITIVE LOGITS
     corporate
    3.41
     Corporate
    2.64
     corporations
    2.16
     corporation
    1.90
    porate
    1.76
     corpor
    1.55
     shareholder
    1.52
    Companies
    1.52
     Corpor
    1.48
     multinational
    1.44
    Activations Density 0.012%

    No Known Activations