© 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. 1670
    Prev
    Next
    INDEX
    Explanations

    negations, specifically cases where statements are negated

    oai_token-act-pair · gpt-3.5-turbo

    phrases involving the concept of negation or denial

    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
    ngth
    -0.72
    isal
    -0.70
    glers
    -0.70
    baugh
    -0.69
    ierrez
    -0.69
    lace
    -0.67
    vant
    -0.67
    rieve
    -0.66
    ãĤ©
    -0.66
     ³³
    -0.65
    POSITIVE LOGITS
     outright
    1.15
     downright
    1.06
     necessarily
    0.90
     entirely
    0.88
     impossible
    0.86
     nonexistent
    0.85
     fatally
    0.83
     altogether
    0.82
     permanently
    0.78
     wholly
    0.78
    Activations Density 0.057%

    No Known Activations