© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 20102
    Prev
    Next
    INDEX
    Explanations

    mentions of things being reduced or made lower

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "lowered" and its variations

    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
    agent
    -0.68
    Writ
    -0.67
    pixel
    -0.66
    archive
    -0.66
     prolific
    -0.65
    amia
    -0.64
     prose
    -0.64
    nom
    -0.63
    atom
    -0.63
    ZA
    -0.62
    POSITIVE LOGITS
     lowering
    2.96
     lowered
    2.84
     lowers
    2.36
    lower
    1.77
    Lower
    1.74
     lower
    1.57
     Lower
    1.53
    Higher
    1.35
     LOW
    1.25
    Low
    1.17
    Activations Density 0.019%

    No Known Activations