© 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. 9-RES-JB
    11. 23906
    Prev
    Next
    INDEX
    Explanations

    narrative descriptions or quotes within a written piece

    oai_token-act-pair · gpt-3.5-turbo

    significant moments or elements in narratives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     awaits
    -0.70
    yet
    -0.70
     currently
    -0.67
    currently
    -0.67
     Currently
    -0.66
    Redditor
    -0.64
     Appears
    -0.63
     varies
    -0.61
    shall
    -0.61
     greets
    -0.60
    POSITIVE LOGITS
     [
    1.06
     ['
    1.05
     naïve
    0.88
     reluct
    0.85
     naive
    0.85
     '[
    0.84
     hindsight
    0.82
     realization
    0.81
     retrospect
    0.81
    ."[
    0.80
    Activations Density 0.796%

    No Known Activations