© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 3-RES-JB
    11. 18518
    Prev
    Next
    INDEX
    Explanations

    references to a specific object - a "straw" (either literal or metaphorical)

    oai_token-act-pair · gpt-3.5-turbo

    the term "straw" in various contexts, indicating its significance in the document

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    olon
    -0.80
    ervation
    -0.75
    mercial
    -0.71
    apse
    -0.71
     notor
    -0.70
    ogue
    -0.69
    ccording
    -0.68
    ECT
    -0.67
     Reloaded
    -0.67
    cients
    -0.67
    POSITIVE LOGITS
     straw
    1.39
     Straw
    0.94
    backs
    0.92
    mere
    0.89
    weights
    0.88
    weight
    0.86
    poll
    0.86
    stad
    0.83
    \\\\\\\\
    0.79
    \\\\\\\\\\\\\\\\
    0.78
    Activations Density 0.009%

    No Known Activations