© 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. 16936
    Prev
    Next
    INDEX
    Explanations

    specific abbreviations commonly associated with technology and engineering, particularly "TE"

    oai_token-act-pair · gpt-3.5-turbo

    references to tight ends in football

    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
    Nap
    -0.71
     Glacier
    -0.68
     Magn
    -0.67
    avia
    -0.67
     McInt
    -0.66
     Lauder
    -0.66
    Magn
    -0.66
    orious
    -0.65
    aunch
    -0.62
     disadvant
    -0.61
    POSITIVE LOGITS
     TE
    3.76
    TE
    1.76
     Te
    1.43
     WR
    1.38
    Te
    1.30
     FB
    1.28
     KE
    1.27
     DE
    1.25
     DT
    1.24
     TA
    1.19
    Activations Density 0.013%

    No Known Activations