© 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. 11-RES-JB
    11. 2312
    Prev
    Next
    INDEX
    Explanations

    contractions with the word "not" and variations of the word "have."

    oai_token-act-pair · gpt-3.5-turbo

    negative contractions, particularly "can't" and "don't"

    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
    ature
    -0.67
    CI
    -0.65
    cycl
    -0.61
    aldi
    -0.60
    ENE
    -0.59
    xy
    -0.59
    76561
    -0.57
    atered
    -0.57
    ogue
    -0.57
    ours
    -0.57
    POSITIVE LOGITS
     resist
    1.08
     conceive
    0.96
     muster
    0.94
     withstand
    0.89
     afford
    0.87
     persuade
    0.87
     disagree
    0.86
     reconcile
    0.86
     restrain
    0.85
     possibly
    0.84
    Activations Density 0.049%

    No Known Activations