© 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. 8134
    Prev
    Next
    INDEX
    Explanations

    mentions of technology-related terms in a text

    oai_token-act-pair · gpt-3.5-turbo

    mentions of technology-related terms

    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
    Ëľ
    -0.78
     Seym
    -0.78
    chwitz
    -0.74
    Written
    -0.72
    lli
    -0.70
     Byrd
    -0.68
    deen
    -0.68
     Flavoring
    -0.67
     redeemed
    -0.66
    Ë
    -0.66
    POSITIVE LOGITS
     tech
    3.60
    tech
    2.31
    Tech
    2.29
     Tech
    2.28
     technology
    2.14
    technology
    1.78
    Technology
    1.76
     techn
    1.66
     Silicon
    1.66
     Technology
    1.61
    Activations Density 0.006%

    No Known Activations