© 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. 11-RES-JB
    11. 12272
    Prev
    Next
    INDEX
    Explanations

    apology-related language and actions, including apologies, plea deals, and acknowledgments

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "apology" and related concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    fixes
    -0.79
    afety
    -0.77
    effic
    -0.73
    umbers
    -0.71
    addons
    -0.71
    licts
    -0.71
    control
    -0.70
     ancest
    -0.70
    effects
    -0.70
    iders
    -0.70
    POSITIVE LOGITS
     letter
    1.53
     Letter
    1.20
     memo
    1.17
     apology
    1.16
     gesture
    1.16
     plea
    1.13
     memorandum
    1.13
     message
    1.11
     statement
    1.11
     ceremony
    1.11
    Activations Density 0.219%

    No Known Activations