© 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. 8-RES-JB
    11. 2529
    Prev
    Next
    INDEX
    Explanations

    words related to organizations or associations, particularly ones with acronyms like "BA" or "CA"

    oai_token-act-pair · gpt-3.5-turbo

    references to beer associations or organizations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lings
    -0.98
    ician
    -0.84
    meyer
    -0.83
    ness
    -0.82
    mans
    -0.80
    tie
    -0.80
    sson
    -0.79
    nir
    -0.79
    aries
    -0.79
    nesses
    -0.77
    POSITIVE LOGITS
    VE
    1.46
    BILITY
    1.32
    UGH
    1.28
    ZE
    1.25
    BIP
    1.23
    KE
    1.23
    UTH
    1.21
    HAHAHAHA
    1.20
    HAHA
    1.20
    BIL
    1.18
    Activations Density 0.068%

    No Known Activations