© 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. 1302
    Prev
    Next
    INDEX
    Explanations

    specific names and titles of individuals in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    references to rankings, titles, or positions in a competitive context

    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
    etheless
    -1.00
     guiName
    -0.79
    conservancy
    -0.66
    âĶĢâĶĢâĶĢâĶĢ
    -0.64
    20439
    -0.63
     tremend
    -0.61
     glim
    -0.60
    ²¾
    -0.60
     streng
    -0.59
    Ô
    -0.59
    POSITIVE LOGITS
    )
    2.08
    )"
    2.04
    ),
    1.99
    )'
    1.92
    ),"
    1.88
    )",
    1.87
    ).
    1.86
    ):
    1.86
    )."
    1.86
    )—
    1.83
    Activations Density 0.428%

    No Known Activations