© 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. 15132
    Prev
    Next
    INDEX
    Explanations

    statements of apology

    oai_token-act-pair · gpt-3.5-turbo

    references to apologies

    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
    Phase
    -0.72
    ield
    -0.67
     stacked
    -0.65
     closely
    -0.65
     favored
    -0.64
    VT
    -0.63
     Variable
    -0.63
     Phase
    -0.63
     combinations
    -0.61
    hare
    -0.61
    POSITIVE LOGITS
     apology
    3.80
     apologies
    2.77
     apologizing
    2.44
     apologized
    2.35
     apologise
    2.30
     apologize
    2.27
     apologised
    2.26
     apolog
    1.96
     remorse
    1.73
     acknowledgement
    1.61
    Activations Density 0.027%

    No Known Activations