© 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. 0-RES-JB
    11. 15337
    Prev
    Next
    INDEX
    Explanations

    references to high-ranking officials or figures within a structured organization

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "cardinals."

    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
     hearing
    -0.78
     Mek
    -0.67
     Hazard
    -0.67
     Ples
    -0.64
     Sparkle
    -0.61
     overhe
    -0.60
     Mechan
    -0.58
     hear
    -0.57
     dwarf
    -0.56
     aide
    -0.55
    POSITIVE LOGITS
    inals
    5.04
    inal
    2.50
    INAL
    2.47
    inally
    2.05
    inates
    1.37
    inators
    1.26
    inate
    1.21
    inis
    1.11
    inating
    1.11
    iaries
    1.07
    Activations Density 0.012%

    No Known Activations